- stunnel-users - stunnel.org

A Calypso user needs your help with stunnel
by John Bryant 02 Sep '08

02 Sep '08

Dear folks: I am trying to install stunnel so that I can have an SSL capability with Calypso. I was guided to the stunnel.org site by instructions at: http://www.windowsbbs.com/courier-email-client/35188-ssl-connection-really-…. (the instructions that I am referring to are duplicated below, under the first line of asterisks - they mention later versions of stunnel, but do not update their instructions for them adequately) I could not understand the instructions at your site that told about modifying the conf file with the LATEST version of stunnel.exe, which I did download and install and that's as far as I got - couldn't understand the options as clearly as I had the others, so uninstalled it.) (Tried downloading the OLD version, 4.05 exe, mentioned in the instructions and the two dll files,and making all the changes to THAT conf.txt file, etc, but THAT did not work because when I ran the shortcut to the conf.txt file that I had put in the Start folder, the computer gave me, when connected, a "cannot find 1 1/4 dll" message, and then an "illegal op" and a "cannot run due to restrictions on your computer" message (when not connected, just the latter two). I know you don't want to hear about old versions but it was logical for me to try since I could not figure out how to use the latest version.) So what do I do now? Thanks! Lenora (do not give up on me, I have computers in my background, etc. etc) ************** (Excerpts from the thread at the url listed above) http://www.windowsbbs.com/courier-email-client/35188-ssl-connection-really-… A week ago I provided instructions for using stunnel to provide SSL functionality for Calypso/Courier - in case your email provider requires SSL, which Calypso and Courier do not yet support. Those instructions were correct if your email provider uses the dedicated SSL ports - 465 for smtps, 995 for pop3s, 993 for imap4s. They were incomplete if your email provider uses the standard ports - 25 for smtp, 110 for pop3, [N/A - 143 for imap4]. Normally when the SSL ports are being used the entire conversation is encrypted with SSL. stunnel will initiate the session with the standard SSL protocol. However, if the standard ports are being used the session is initiated without SSL and then switches to SSL using an smtp or pop3 specific protocol (STARTTLS or STLS). stunnel needs to be told this, or it wont work on the standard ports. For the standard ports an extra line is required in the configuration file. To use stunnel do the following: 1) download the following files from http://www.stunnel.org/download/binaries.html stunnel-4.05.exe (version may be higher when you go there) libssl32.dll libeay32.dll and put them into a directory such as C:\Program Files\Stunnel\ 2) create a shortcut to stunnel-4.05.exe in the same directory [in WinXP do this by right clicking on it and selecting Create Shortcut] 3) edit the shortcut by right clicking on it and selecting Properties Modify the Target to look like this "C:\Program Files\Stunnel\stunnel-4.05.exe" stunnel-conf.txt (Optional) Rename the shortcut to remove the "Shortcut to ". 4) create the file stunnel-conf.txt in the same directory, and put the following lines in it: ----starting with next line---- # GLOBAL OPTIONS client = yes output = stunnel-log.txt debug=4 taskbar = yes # SERVICE-LEVEL OPTIONS [POP3 (ISP name)] accept = 127.0.0.1:108 connect = myISPs.POP3server.net:995 [SMTP (ISP name)] accept = 127.0.0.1:107 connect = myISPs.SMTPserver.net:465 ----ending with previous line---- Change the server names (before colons), and perhaps ports (after colons), to be correct for your ISP. Change ISP name too. Save and exit. 4b) if your email provider uses standard ports for either smtp or pop3, make either or both of the following changes to your stunnel-conf.txt file: [POP3 (ISP name)] accept = 127.0.0.1:108 connect = myISPs.POP3server.net:110 protocol = pop3 [SMTP (ISP name)] accept = 127.0.0.1:107 connect = myISPs.SMTPserver.net:25 protocol = smtp Once again, change the server names (before colons) to be correct for your ISP. Change ISP name too. Save and exit. 4c) if you have two email providers requiring SSL you can add a second set of services like so: [POP3 (2nd ISP name)] accept = 127.0.0.1:106 connect = my2ndISPs.POP3server.net:995 [SMTP (2nd ISP name)] accept = 127.0.0.1:105 connect = my2ndISPs.SMTPserver.net:465 Each email provider requires its own two ports (e.g. 108+107, 106+105), but 2 email accounts at a single provider can share the same pair of ports. However, if an email provider only uses SSL for one of smtp or pop3, you only need one of the two service sections. Once again, change the server names (before colons) to be correct for your ISP. Change ISP name too. Save and exit. 5) put a copy of that shortcut into your Start Menu's Programs/Startup folder [normally, by dragging shortcut over Start button, over Programs, over Startup, then drop it in Startup list.] 6) edit your Calypso/Courier Account Properties to change your mail servers. Incoming: Port 108 Server 127.0.0.1 Outgoing: Port 107 Server 127.0.0.1 [this is done differently for Calypso and Courier, due to Courier's new approach to SMTP servers] If an email provider only uses SSL for one of smtp or pop3, you only need to change one of the two mail servers. If you set up more than one email provider for SSL, change each account's properties to the corresponding ports. Now when you check or send email Calypso/Courier will communicate with stunnel, which will connect to your ISP's mail servers via SSL. The shortcut in Startup will rerun stunnel each time you reboot. You should just double-click on the shortcut in the program folder to run stunnel the first time. Hope this helps. Greg ************* Later addition (later on same thread): I just reinstalled all of my computer and I forgot to backup my sTunnel config! Anyway, the instructions have changed a bit as of sTunnel version 4.12... sTunnel is now distributed as an installer EXE, and the auxiliary DLLs are installed by default. The only step that really changes is step 1. All you have to do is to download the sTunnel installer and run it. There is no need to download the other files. The rest stays pretty much the same. ********** Question from somebody: First of all, I want to thank you for posting this very useful 'how to' -- this is what foruming is all about I have two questions (for anyone): 1. When using Stunnel, does it completely encrypt the user name, password, and the contents of the email, or just the contents? 2. The Stunnel website (www.stunnel.org) mentions that you'd also need the OpenSSL Libraries to make it work. But I did not download them, and still managed to make the Stunnel work with Courier. Am I (still) missing something? ************ Answer: 1. stunnel establishes an ssl tunnel (hence the name) between your computer and the email server. So every byte travelling between the two is encrypted. It's like establishing a VPN between your computer and the server. 2. As gmagana posted at the start of the month, they seem to have enhanced the installation process - such that you don't need to worry about the extra files. Isn't open source great! ********* Final comment: Looks like I can finally use Courier with Google's Gmail (http://gmail.google.com) to send/receive encrypted email. (End of excerpts)

2 1

Re: [stunnel-users] failed to build stunnel 4.25 under VS 2005
by Michal Trojnara 29 Aug '08

29 Aug '08

On Fri, 29 Aug 2008 00:00:56 +0800, "Ericdai" <pcix133(a)hotmail.com> wrote: > I try to build Stunnel 4.25 under VS2005 ,but failed Today I updated stunnel, so it can be built with Visual C++ 2008 (known also as Visual C++ 9.0). My plan is to release stunnel 4.26 next week. > prototypes.h(428) : error C2081: 'socklen_t' : name in formal > parameter list illegal I had no problem with socklen_t. Maybe it's specific to the old version of Visual C++. > I also suggest build a windows client only version > n It only can be running on windows > n It only can act as a client > n I think it can reduce the code size and runtime footprice smaller Not really. The difference would be negligible. Best regards, Mike

1 0

failed to build stunnel 4.25 under VS 2005
by Ericdai 28 Aug '08

28 Aug '08

I try to build Stunnel 4.25 under VS2005 ,but failed cl -MD -W3 -Ox -O2 -Ob2 -Gs0 -GF -Gy -GL -nologo -I"..\..\ssl\inc32" -DUSE_WIN32 -D_CRT_SECURE_NO_DEPRECATE -D_CRT_NONSTDC_NO_DEPRECATE -DHAVE_GETADDRINFO -DHAVE_GETNAMEINFO -D_MBCS /c stunnel.c stunnel.c prototypes.h(428) : error C2081: 'socklen_t' : name in formal parameter list illegal stunnel.c(206) : error C2065: 'socklen_t' : undeclared identifier stunnel.c(206) : error C2146: syntax error : missing ';' before identifier 'addrlen' stunnel.c(206) : error C2065: 'addrlen' : undeclared identifier I use nmake -f vc.mak to do it Anyone has build it successfully ,pls help me ! Another way ,is there a stunnel solution or project file ? I also suggest build a windows client only version n It only can be running on windows n It only can act as a client n I think it can reduce the code size and runtime footprice smaller Thanks and best regards Eric dai

1 0

Re: [stunnel-users] In chrooted environment system times defaults to UTC
by Ivan Lezhnjov Jr. 22 Aug '08

22 Aug '08

Friday 22 August 2008 you wrote: > I noticed that when running stunnel from within the jail it logs its > messages with syslong using UTC. > > My question is how would one reconcile timezone of chrooted stunnel with a > system one? > > The issue illustrated: > > Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: stunnel 4.17 on > i686-pc-linux-gnu with OpenSSL 0.9.8h 28 May 2008 > Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: Threading:PTHREAD > SSL:ENGINE Sockets:POLL,IPv4 > Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: 500 clients > allowed Aug 22 12:29:40 xerxes stunnel: LOG5[26570:3086773168]: 5101 > connected from 127.0.0.1:1469 > Aug 22 09:29:40 xerxes stunnel: LOG5[26570:3086773168]: VERIFY OK: > depth=0, /C=UA/ST=ARC/L=Simferopol/O=HOMELAB/CN=sega.localnet > > xerxes' localtime is GMT+3 and sega's too, but stunnel runs on sega in > chrooted environment and thus uses UTC (if it's not chrooted it reports > localtime to syslog facility). > > Any ideas? Jee that was simple! cp /etc/localtime /var/lib/stunnel/etc/ (on each machine where stunnel runs) and you're done :^) -- Ivan Lezhnjov Jr. Europe, Ukraine, Simferopol Running Source Mage GNU/Linux, kernel version 2.6.24 build #5 +----------------------------------------------------------------------+ Key ID 0x5811D90C Key Fingerprint 2A52 5C8C 38BE C04F D8DE A169 19E2 E49A 5811 D90C Use GPG Exercise Your Right To Privacy

1 0

In chrooted environment system times defaults to UTC
by Ivan Lezhnjov Jr. 22 Aug '08

22 Aug '08

I noticed that when running stunnel from within the jail it logs its messages with syslong using UTC. My question is how would one reconcile timezone of chrooted stunnel with a system one? The issue illustrated: Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: stunnel 4.17 on i686-pc-linux-gnu with OpenSSL 0.9.8h 28 May 2008 Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv4 Aug 22 12:29:40 xerxes stunnel: LOG5[26569:3083679424]: 500 clients allowed Aug 22 12:29:40 xerxes stunnel: LOG5[26570:3086773168]: 5101 connected from 127.0.0.1:1469 Aug 22 09:29:40 xerxes stunnel: LOG5[26570:3086773168]: VERIFY OK: depth=0, /C=UA/ST=ARC/L=Simferopol/O=HOMELAB/CN=sega.localnet xerxes' localtime is GMT+3 and sega's too, but stunnel runs on sega in chrooted environment and thus uses UTC (if it's not chrooted it reports localtime to syslog facility). Any ideas? -- Ivan Lezhnjov Jr. Europe, Ukraine, Simferopol Running Source Mage GNU/Linux, kernel version 2.6.24 build #5 +----------------------------------------------------------------------+ Key ID 0x5811D90C Key Fingerprint 2A52 5C8C 38BE C04F D8DE A169 19E2 E49A 5811 D90C Use GPG Exercise Your Right To Privacy

1 0

Certificates, Servers & Clients. Clarify this please.
by Ivan Lezhnjov Jr. 22 Aug '08

22 Aug '08

Now that I've read a bunch of articles and have seen a pile of examples of how pairs of a certificate and a private key can be created I must say they all suggest different approaches and techniques to creating certificates and ultimately bewildered me. Could anyone please explain as simply as possible, in layman's terms what exactly must be done? Let's talk about simple scenario with two hosts (A & B) involved, both running stunnel. Let us assume host A is the server and host B is the client. cert = CAfile = must be created. And we use verify level 3 (verify = 3) So, what exactly do we do now? And what happens in the dialog between these two machines? -- Ivan Lezhnjov Jr. Europe, Ukraine, Simferopol Running Source Mage GNU/Linux, kernel version 2.6.24 build #5 +----------------------------------------------------------------------+ Key ID 0x5811D90C Key Fingerprint 2A52 5C8C 38BE C04F D8DE A169 19E2 E49A 5811 D90C Use GPG Exercise Your Right To Privacy

1 0

Can't start stunnel (dies mysteriously)
by Ivan Lezhnjov Jr. 19 Aug '08

19 Aug '08

Hello there I run stunnel version 4.21 and openssl version 0.9.8h. In my setup I aim to create a tunnel to send log files through to a centralized loghost. Host A (loghost) runs Source Mage GNU/Linux (with stunnel 4.21, openssl 0.9.8h) and host B runs Slackware 11 system (with stunnel 4.17, openssl 0.9.8h). The problem is that stunnel starts on host B but won't start on host A. Here's the snippet of /var/log/messages that records stunnel start-up event: root@sega:/home/users/ilj % stunnel && tail /var/log/messages -n 3 Aug 19 16:17:37 sega stunnel: LOG5[29146:3082634944]: stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8h 28 May 2008 Aug 19 16:17:37 sega stunnel: LOG5[29146:3082634944]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Aug 19 16:17:37 sega stunnel: LOG5[29146:3082634944]: 500 clients allowed I've actually been trying to run stunnel for the first time ever so I don't really know if this output confirms that everything is alright but judging solely from what is put in log file it seems stunnel has started successfully. Unfortunately, this isn't quite true root@sega:/home/users/ilj % ps ax | grep stun 29618 pts/4 R+ 0:00 grep stun So, this is the problem I have. It doesn't start on host A. Host A has the following stunnel configuration: root@sega:/home/users/ilj % cat /etc/stunnel/stunnel.conf cert = /etc/stunnel/syslog-ng-server.pem CAfile = /etc/stunnel/syslog-ng-client.pem verify = 3 [5101] accept = 217.117.75.2:5101 connect = 127.0.0.1:1999 Host B sports a little bit different configuration file: root@xerxes:~ % cat /etc/stunnel/stunnel.conf client = yes cert = /etc/stunnel/syslog-ng-client.pem CAfile = /etc/stunnel/syslog-ng-server.pem verify = 3 [5101] accept = 127.0.0.1:1999 connect = 217.117.75.2:5101 On both hosts (A & B) I run stunnel standalone as root. > 4. Output of "stunnel -f -D 7 <your-parameters>". Erm.. it doesn't seem to work for me. There's no -D parameter according to man stunnel. The following information below is about host A only: root@sega:/home/users/ilj % stunnel -version stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8h 28 May 2008 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /usr/var/run/stunnel/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = AES:ALL:!aNULL:!eNULL:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none root@sega:/home/users/ilj % uname -a Linux sega 2.6.24 #5 SMP PREEMPT Wed May 7 18:22:29 EEST 2008 i686 GNU/Linux root@sega:/home/users/ilj % gaze from /usr/lib/libc.a glibc-2.7:/usr/lib/libc.a root@sega:/home/users/ilj % gcc -v Reading specs from /usr/lib/gcc/i686-pc-linux-gnu/4.2.3/specs Target: i686-pc-linux-gnu Configured with: /usr/src/gcc-4.2.3/configure --host=i686-pc-linux-gnu --prefix=/usr --infodir=/usr/share/info --mandir=/usr/share/man --enable-threads=posix --with-system-zlib Thread model: posix gcc version 4.2.3 root@sega:/home/users/ilj % gaze installed | grep openssl openssl:20080615:installed:0.9.8h -- Ivan Lezhnjov Jr. Europe, Ukraine, Simferopol Running Source Mage GNU/Linux, kernel version 2.6.24 build #5 +----------------------------------------------------------------------+ Key ID 0x5811D90C Key Fingerprint 2A52 5C8C 38BE C04F D8DE A169 19E2 E49A 5811 D90C Use GPG Exercise Your Right To Privacy

1 0

Disable libwrap at runtime?
by Jans, Kert 16 Aug '08

16 Aug '08

Hi, Is it possible to disable libwrap support at runtime? I've reviewed the archives and documentation and cannot determine if this is possible. I know it is possible to compile-out support for libwrap by using: configure --disable-libwrap Is compiling-out support the only way? I'd like to disable support in a pre-build version of stunnel. Regards, Kert

2 1

Re: [stunnel-users] Mutual SSL Authentication
by Christopher Hunt 15 Aug '08

15 Aug '08

Ah ha - "verify = 3" - thanks! On 15/08/2008, at 9:12 AM, Brian Hatch wrote: >> > Mutual authentication uses the verify option on both > ends, rather than just on one end. Both sides verify, > and both sides present a certificate.

2 1

Re: [stunnel-users] Mutual SSL Authentication
by Christopher Hunt 14 Aug '08

14 Aug '08

On 14/08/2008, at 8:00 PM, Michal.Trojnara(a)mobi-com.net wrote: > Sure. Stunnel supports mutual SSL authentication. > Please read the fine manual for details. Hi Mike, Thanks for the reply. I did quite a lot of trawling on the web, your website and the stunnel man pages and couldn't find any guidance on mutual ssl authentication configuration. Are you able to give me a pointer, perhaps a URL to something that contains an example? Thanks again. Cheers, -C

1 0

Stunnel connection drops after four http requests
by Renner, Christopher R 13 Aug '08

13 Aug '08

With quite a bit of help from a list member here, I have stunnel 4.25 running on OSX 1.5.4 (PHP development environment). Problem is I can only get four http requests to the server before the connection is dropped. Stunnel is still running in the background.

1 0

Mutual SSL Authentication
by Christopher Hunt 13 Aug '08

13 Aug '08

Hi there, Does stunnel have any in-built support for mutual SSL authentication? If not are there plans to incorporate support? Cheers, -C

2 1

question
by Ted Ehrlich 12 Aug '08

12 Aug '08

1 0

OSX 10.5 install?
by Renner, Christopher R 11 Aug '08

11 Aug '08

> >> Hi, >> >> New list member here. I¹m going through the archives now, and have Googled >> my way to exhaustion, looking for up-to-date instructions to install stunnel >> on a Mac desktop. >> >> I do PHP web application development and need to be able to connect to a NAS >> space via stunnel on my desktop dev environment in order to simulate the live >> environment. >> >> Does anyone 1) have a link to current OSX install instructions, or 2) care to >> share such instructions with me off the top of your head? >> >> I tried working through this darwinports page ( >> http://stunnel.darwinports.com/) but 1) it sends me to Macports.org for the >> .dmg file thereby installing macports and not darwinports....and after that >> the ³port² command is unrecognized in terminal when trying to install >> stunnel. >> >> >> Thanks. >>

2 1

Socket connection status
by Jeffrey Buck 08 Aug '08

08 Aug '08

Hi, I'm running stunnel 4.04 on ia64-redhat-linux-gnu PTHREAD+LIBWRAP with OpenSSL 0.9.7a Feb 19 2003. I have basically taken a generic TCP interface and implemented a STUNNEL client/server setup to allow the connection to transfer data using SSL. I have an issue with checking the status of the connection when a connection attempt is made. We are using ObjectSpace libraries for all of our tcp related connections so I don't have the low level calls that are actually being made, but here is a brief description of what is happenning.... We create an os_tcp_socket object and call the connect_to() method. This establishes a connection to the stunnel client on my local machine....which in turn makes the connection the the stunnel server on the remote machine. The problem I'm am having is when I can't get to the remote machine. After the connect_to() call, we check the status of the socket and unless I sleep for 5-10 seconds, it always seems to think that the connection is ok...which kind-of makes sense because it is able to connect to the stunnel client port since it is local...but in fact, if there isn't access to the remote machine, eventually stunnel will close that initial connection to that socket (that where the 5-10 second delay seems to help). The status I really want is weather or not the entire connection was successful through stunnel to the remote machine. Is there a better way to do this other than putting a sleep() in. Right now I have a 10 second sleep in between the connect_to() call and the status check on the local/client stunnel TCP port. That seems to be enough time for stunnel to attept to complete the connection and kill the connection to it's local/client port so that when I check the status of it, it comes back bad. If this is the best way to do it...what is the amount of time needed for stunnel to attempt to complete it's connection to the remote machine? Thanks! Jeff

2 3

Yahoo SSL
by Bernie Murciano 29 Jul '08

29 Jul '08

1 0

Re: [stunnel-users] help with stunnell
by Ludolf Holzheid 28 Jul '08

28 Jul '08

On Thu, 2008-07-24 23:37:22 -0700, vinodsreehari(a)yahoo.com wrote: > [..] > > I configured stnnel in one server as server mode ,for accepting > connection on 8070 ,and connect to 80 . It works only if i specify > https://ipaddress:8070 This is because the default port for https is 443. If you'd configure stunnel to accept on port 443, `https://ipaddress' would work. > I have configured one client server also to accept connection on 80 > and redirect to teh 8070 of server. I want to know that is that teh > way its working . If there is a stunnel instance on the client machine accepting connections on port 80 and connecting to the web server at the https port (default or not), a user could specify `http://localhost' to connect to the web server (and the traffic between the client machine and the server would be secured). > And is it possible to implement the same in a live enviroment ,in > which teh webserver will run on different port ,but users dont need > to specify the port number . Yes. As soon as the destination port for the 'first hop' of a redirected/tunneled connection is the default port for the protocol in question, there is no need to specify the port number. The port numbers of 'intermediate hops' are configuration options for the two stunnel instances (in this case), so the user dosn't have to care about them. HTH, Ludolf -- --------------------------------------------------------------- Ludolf Holzheid Tel: +49 621 339960 Bihl+Wiedemann GmbH Fax: +49 621 3392239 Floßwörthstraße 41 e-mail: lholzheid(a)bihl-wiedemann.de D-68199 Mannheim, Germany ---------------------------------------------------------------

1 0

Re: [stunnel-users] Wrong documentation
by Sandeep Kumar 25 Jul '08

25 Jul '08

On Thu, Jul 24, 2008 at 9:39 PM, Brian Hatch <bri(a)stunnel.org> wrote: > In the neigborhood of 2008-07-24 09:40 +0530, Sandeep Kumar mentioned: > > > Look at this page > http://www.stunnel.org/faq/stunnel.html#configuration_file > > In service level options section, under the explanation of "CRLpath" it > is > > said that all files in it should be of the form XXXXXXXX.0. But my > stunnnel > > setup worked only when i names as XXXXXXXX.r0 instead of XXXXXXXX.0. > > What version of openssl are you running? stunnel 4.15 with OpenSSL 0.9.8b > > > > -- > Brian Hatch "Have you asked Kirk if > Systems and he'd like the job?" > Security Engineer "You can't hide from me! > http://www.ifokr.org/bri/ I will find you. > > Every message PGP signed > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.2.1 (GNU/Linux) > > iD8DBQFIiKlFVkMj8/ymYEsRAmKxAJ9CTLp0Xi8n1VxdFup/kBUuZe6scwCfcvyu > WQG8BlfrUV8RcP31/ZUEbJk= > =EBgc > -----END PGP SIGNATURE----- > > -- Sandeep Kumar http://students.iiit.ac.in/~sandeep_kr

1 0

Help on configuring Stunnel
by Patel Dippen-CDP054 24 Jul '08

24 Jul '08

I downloaded the pre-compiled Windows version of Stunnel (stunnel-4.25-installer.exe). We need to deploy this on 2 machines A and B. The way it works is: - B connects to A - then A connects to B How should I configure Stunnel on machine A, since, in the first case A is the Server and in the 2nd scenario A becomes the Client. Questions on deploying Stunnel on machine A: - Can I run just 1 instance of Stunnel on machine A? It seems I cannot since I have to set the "client = yes" in the 2nd scenario - Do I need to run multiple instances, 1 running as the Server and the 2nd as the Client? Here is the config file I used on the machine A: **************************** cert = stunnel.pem socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 [iPerfS] accept = 5010 connect = 127.0.0.1:5001 [iPerfS] accept = 5002 connect = 12.50.0.25:5020 **************************** Here is the config file I used on the machine B: **************************** cert = stunnel.pem socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 [iPerfC] accept = 5001 connect = 12.50.0.26:5010 [iPerfS] accept = 5020 connect = 127.0.0.1:5002 **************************** Regards, Dippen

1 0

Stunnel 4.24 patch to allow stunnel to search for the OCSP responder within an X.509 certificate's AIA extension.
by Bruce Keats 24 Jul '08

24 Jul '08

Hi all, This patch also fixes a problem with stunnel accepting an "unknown" response as "good" and allows the connection to be setup. An OCSP responder can return one of three responses (good, revoked, unknown), however the code would accept "unknown" as good. This patch fixes that problem and treats an "unknown" response as a failure. Stunnel can query an OCSP responder to determine the revocation status of a certificate when stunnel is verifying the certificate. The code works great, however it is uses a preconfigured OCSP responder. The OCSP responder's IP/hostname is specified in the stunnel configuration file as a URI. I found this rather restrictive, so I added code to extract the OCSP responder's URI from the certificate. As per RFC 3280, one of the valid extensions for X.509 certificate is the Authority Information Access (AIA) where the "accessMethod" can be "is-ad-ocsp" to identify the OCSP responders location as per RFC2560. There can be more than one OCSP responder listed in the AIA because it is a sequence. Each location is queried in the order they appear within the certificate until either a "good" or "revoked" response is received. If an "known" response or a failure then the next OCSP responder is checked. If all OCSP responder queries result in failure then the validation also fails.. A new configuration option "OCSPaia" was added. If the "OCSPaia=yes" then the stunnel will look for the OCSP responder(s) within the certificate's AIA. The locally configured "OCSP=<url>" overrides the OCSPaia option. Bruce

1 0

Wrong documentation
by Sandeep Kumar 24 Jul '08

24 Jul '08

Look at this page http://www.stunnel.org/faq/stunnel.html#configuration_file In service level options section, under the explanation of "CRLpath" it is said that all files in it should be of the form XXXXXXXX.0. But my stunnnel setup worked only when i names as XXXXXXXX.r0 instead of XXXXXXXX.0. -- Sandeep Kumar http://students.iiit.ac.in/~sandeep_kr

1 0

help with stunnel port forwarding
by vinod bhasi 23 Jul '08

23 Jul '08

Hi, Please help me with stunnel port forwarding . My apache is running at port 8000 .I want this to be forwarded to 80 so that normal users dont need to specify the port number . Can anyone help me with the configuration in both client and server side .Im confused as stunnel wont run with same port used by apache. Any help will be appreciated Thanks in advancve Vinod Bhasi Did you know? You can CHAT without downloading messenger. Go to http://in.webmessenger.yahoo.com/

2 1

Re: [stunnel-users] stunnel-users Digest, Vol 48, Issue 13
by Patrick Dung 21 Jul '08

21 Jul '08

Message: 3 Date: Mon, 21 Jul 2008 11:34:22 +0200 From: Micha? Trojnara <Michal.Trojnara(a)mobi-com.net> Subject: Re: [stunnel-users] Problem with Solaris 10u5 x86 and Stunnel 4.25 To: stunnel-users(a)mirt.net Message-ID: <2a3f79f5f6dd65f2ced02b2e9929368b(a)mirt.net> Content-Type: text/plain; charset="UTF-8" http://stunnel.mirt.net/pipermail/stunnel-users/2008-June/002021.html Mike On Sat, 19 Jul 2008 01:53:17 -0700 (PDT), Patrick Dung <patrick_dkt(a)yahoo.com.hk> wrote: > Hi > > I use this to compile stunnel: > The gcc and openssl is from Solaris SFW. > > # export CC=/usr/sfw/bin/gcc > # ./configure --with-threads=pthread --with-ssl=/usr/sfw > --localstatedir=/var > > After installing and starting stunnel, I found there are mutiple stunnel > processes. > There is a PID file which store one of the stunnel process. > I have tried to kill that PID, stunnel is bring down. > But there other stunnel processes running. > > Any ideas? > > Regards > Patrick > > > > > _______________________________________________ > stunnel-users mailing list > stunnel-users(a)mirt.net > http://stunnel.mirt.net/mailman/listinfo/stunnel-users I see. Thanks for reply. Yesterday, I have tried to compile stunnel with Sun Studio. It got problem linking with libwrap, so I turned off libwrap support. And it compile (without knowing there is a libwrap patch) and there is only one single running stunnel process. Regards Patrick

1 0

stunnel exits silently on startup
by Luis López de Quintana 21 Jul '08

21 Jul '08

Problem: stunnel doesn't run: ldq@ldq:~$ stunnel4 /etc/stunnel.cnf ldq@ldq:~$ ps aux | grep stunnel ldq 15683 0.0 0.0 2332 1392 pts/0 S+ 11:08 0:00 man stunnel ldq 25289 0.0 0.0 1776 608 pts/1 S+ 11:59 0:00 grep stunnel stunnel.cnf: ldq@ldq:~$ cat /etc/stunnel.cnf client = yes debug = debug [nntps] accept = 127.0.0.1:1109 connect = news.giganews.com:443 Version: stunnel 4.140-5ubuntu1 libc6 ldq@ldq:~$ uname -a Linux ldq 2.6.18.5 SMP Thu May 8 01:31:23 UTC 2008 x86_64 GNU/Linux -- LdQ

2 2

Problem with Solaris 10u5 x86 and Stunnel 4.25
by Patrick Dung 19 Jul '08

19 Jul '08

Hi I use this to compile stunnel: The gcc and openssl is from Solaris SFW. # export CC=/usr/sfw/bin/gcc # ./configure --with-threads=pthread --with-ssl=/usr/sfw --localstatedir=/var After installing and starting stunnel, I found there are mutiple stunnel processes. There is a PID file which store one of the stunnel process. I have tried to kill that PID, stunnel is bring down. But there other stunnel processes running. Any ideas? Regards Patrick

1 0

Config Sanity Check
by Caplan, Michael 19 Jul '08

19 Jul '08

Hi There, I am working through an issue with a script that builds a socket connection to stunnel. Stunnel is being used to relay to an Apache web server. I'm building a localhost TCP connection on port 54322, and writing my HTTP data. When reading data back from the socket connection in my script I'm sometimes getting nothing, and other times getting the proper result. I can confirm that the web server is always receiving the request and returning a proper response. Wireshark shows SSL data being transmitted and received, so stunnel is doing its thing. However, the response does not seem to be relayed to my script on the other side of stunnel *sometimes*. My hunch is that I am doing something wrong with my script, but I am hoping that an stunnel guru can rule out any oddities with my stunnel setup. My config is below as well as the output of my log file. Any thoughts? Thanks! Mike cert = connector.pem debug = 7 output = c:\ssl.log ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 [registerserver] client = yes accept =54321 connect =www.int.ddx.com:443 [connector2server] client = yes accept =54322 connect =asd.int.ddx.com:443 [server2connector] accept = 54323 connect = 9000 2008.07.18 21:41:41 LOG7[13648:13748]: Snagged 64 random bytes from C:/.rnd 2008.07.18 21:41:41 LOG7[13648:13748]: Wrote 1024 new random bytes to C:/.rnd 2008.07.18 21:41:41 LOG7[13648:13748]: RAND_status claims sufficient entropy for the PRNG 2008.07.18 21:41:41 LOG7[13648:13748]: PRNG seeded successfully 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate loaded 2008.07.18 21:41:41 LOG7[13648:13748]: Key file: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Private key loaded 2008.07.18 21:41:41 LOG7[13648:13748]: SSL context initialized for service registerserver 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate loaded 2008.07.18 21:41:41 LOG7[13648:13748]: Key file: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Private key loaded 2008.07.18 21:41:41 LOG7[13648:13748]: SSL context initialized for service connector2server 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate loaded 2008.07.18 21:41:41 LOG7[13648:13748]: Key file: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Private key loaded 2008.07.18 21:41:41 LOG7[13648:13748]: SSL context initialized for service server2connector 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Certificate loaded 2008.07.18 21:41:41 LOG7[13648:13748]: Key file: connector.pem 2008.07.18 21:41:41 LOG7[13648:13748]: Private key loaded 2008.07.18 21:41:41 LOG7[13648:13748]: SSL context initialized for service test 2008.07.18 21:41:41 LOG5[13648:13748]: stunnel 4.25 on x86-pc-mingw32-gnu with OpenSSL 0.9.8h 28 May 2008 2008.07.18 21:41:41 LOG5[13648:13748]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2008.07.18 21:41:46 LOG7[13476:13096]: Snagged 64 random bytes from C:/.rnd 2008.07.18 21:41:47 LOG7[13476:13096]: Wrote 1024 new random bytes to C:/.rnd 2008.07.18 21:41:47 LOG7[13476:13096]: RAND_status claims sufficient entropy for the PRNG 2008.07.18 21:41:47 LOG7[13476:13096]: PRNG seeded successfully 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate loaded 2008.07.18 21:41:47 LOG7[13476:13096]: Key file: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Private key loaded 2008.07.18 21:41:47 LOG7[13476:13096]: SSL context initialized for service registerserver 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate loaded 2008.07.18 21:41:47 LOG7[13476:13096]: Key file: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Private key loaded 2008.07.18 21:41:47 LOG7[13476:13096]: SSL context initialized for service connector2server 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate loaded 2008.07.18 21:41:47 LOG7[13476:13096]: Key file: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Private key loaded 2008.07.18 21:41:47 LOG7[13476:13096]: SSL context initialized for service server2connector 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Certificate loaded 2008.07.18 21:41:47 LOG7[13476:13096]: Key file: connector.pem 2008.07.18 21:41:47 LOG7[13476:13096]: Private key loaded 2008.07.18 21:41:47 LOG7[13476:13096]: SSL context initialized for service test 2008.07.18 21:41:47 LOG5[13476:13096]: stunnel 4.25 on x86-pc-mingw32-gnu with OpenSSL 0.9.8h 28 May 2008 2008.07.18 21:41:47 LOG5[13476:13096]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 Please consider the environment before printing this email. E-mail messages may contain viruses, worms, or other malicious code. By reading the message and opening any attachments, the recipient accepts full responsibility for taking protective action against such code. Henry Schein is not liable for any loss or damage arising from this message. The information in this email is confidential and may be legally privileged. It is intended solely for the addressee(s). Access to this e-mail by anyone else is unauthorized.

1 0

Difference between verify level 2 and 3
by Sandeep Kumar 16 Jul '08

16 Jul '08

Hi, I searched a lot on google, in the archives, and in the FAQ page, but could not find the differences between verify levels 2 and 3, other than the one line explanation given on the FAQ/man page. Specifically, I want to ask: 1. Is verify level 3 same as 2, but asks for client certificates as well? (Seems unlikely but still..) 2. Since verify level 2 also demands CApath/CAfile to be present, then it also is ultimately verifying against locally installed certs, which is supposed to be level 3's behaviour? Thanks & regards, Sandeep Kumar http://web.iiit.ac.in/~sandeep_kr

2 2

stunnel-portforwarding
by vinod bhasi 16 Jul '08

16 Jul '08

Hi All, I am a newbie as far as stunnel is concerned. I need help urgently on how to do port forwarding with Stunnel. Any help is appreciated. Thanks and Regards, Vinod Bhasi Save all your chat conversations. Find them online at http://in.messenger.yahoo.com/webmessengerpromo.php

1 0

SSL with SMTP problem - sTunnel 4.25 Windows
by dscycler 14 Jul '08

14 Jul '08

I am using sTunnel 4.25 for Windows to send and receive e-mail from a GoldMine client. sTunnel has worked well over the years but am having an issue with a particular mail host, Live Office. They have implemented SMTP SSL on port 25. The debug log is as follows: 2008.07.14 13:45:11 LOG7[2136:5356]: smtp accepted FD=224 from 127.0.0.1:1907 2008.07.14 13:45:11 LOG7[2136:5356]: Creating a new thread 2008.07.14 13:45:11 LOG7[2136:5356]: New thread created 2008.07.14 13:45:11 LOG7[2136:4944]: smtp started 2008.07.14 13:45:11 LOG7[2136:4944]: FD 224 in non-blocking mode 2008.07.14 13:45:11 LOG5[2136:4944]: smtp accepted connection from 127.0.0.1:1907 2008.07.14 13:45:11 LOG7[2136:4944]: FD 232 in non-blocking mode 2008.07.14 13:45:11 LOG7[2136:4944]: smtp connecting 64.70.94.196:25 2008.07.14 13:45:11 LOG7[2136:4944]: connect_wait: waiting 10 seconds 2008.07.14 13:45:11 LOG7[2136:4944]: connect_wait: connected 2008.07.14 13:45:11 LOG5[2136:4944]: smtp connected remote server from 127.0.0.1:1908 2008.07.14 13:45:11 LOG7[2136:4944]: Remote FD=232 initialized 2008.07.14 13:45:11 LOG7[2136:4944]: SSL state (connect): before/connect initialization 2008.07.14 13:45:11 LOG7[2136:4944]: SSL state (connect): SSLv3 write client hello A 2008.07.14 13:45:11 LOG7[2136:4944]: SSL alert (write): fatal: handshake failure 2008.07.14 13:45:11 LOG3[2136:4944]: SSL_connect: 1408F10B: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number 2008.07.14 13:45:11 LOG5[2136:4944]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.07.14 13:45:11 LOG7[2136:4944]: smtp finished (0 left) I am looking for help to troubleshoot the connection problem. Is this an sTunnel issue or a mail server issue? Are there conf file settings to address the issue? Thanks, DavidS

1 0

Re: [stunnel-users] stunnel-users Digest, Vol 48, Issue 8
by Pietro Di Primo 10 Jul '08

10 Jul '08

Thank you for reply. I tryied it but it doesn't work because my client run some separated subprocesses and are them that contact the server so, if an extern process try to connect to 127.0.0.1:<the_found_port> while one of that subprocess has finished and the next is not yet started, it will be able to do it and also my subprocess will be pending (i tryed it). Hope i was able to explain the situation. Any suggestion? Thanks in advace > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 9 Jul 2008 13:26:03 +0200 (CEST) > From: "Pietro Di Primo" <pietro.diprimo(a)ct.infn.it> > Subject: [stunnel-users] client-server tunnelling > To: stunnel-users(a)mirt.net > Message-ID: > <2553.193.206.208.201.1215602763.squirrel(a)imapfis.ct.infn.it> > Content-Type: text/plain;charset=utf-8 > > Hi, I have a client server application, and want to tunnel it in this way: > a script select a free port in a random way, write the stunnel.conf file. > with "accept=<the_found_port>" and "connect=<my_server>" and run my client > application passing to 127.0.0.1:<the_found_port> as server address. > The problem is that i need to be sure that no other client or user can > access <the_found_port> while the application is running. > I guess that it should be possible to do it, but i wasn't able to. > > Can anybody help me? > > Thanks > > > -- > --- > > Ing. Pietro Di Primo > INFN - Istituto Nazionale di Fisica Nucleare > Grid Computing Research Area > Via S. Sofia 78 95123 CATANIA (Italy) > Voice : +39 095 378-5446/5517 > Fax. : +39 095 378 5518 > Mobile : +39 320 8591527 > E-mail : pietro.diprimo(a)ct.infn.it > Skype : pietrodiprimo > > > > ------------------------------ > > Message: 2 > Date: Wed, 9 Jul 2008 14:08:23 +0200 > From: Micha? Trojnara <Michal.Trojnara(a)mobi-com.net> > Subject: Re: [stunnel-users] client-server tunnelling > To: stunnel-users(a)mirt.net > Message-ID: <6ba9f1b703149916bb1a61fb68a1e377(a)mirt.net> > Content-Type: text/plain; charset="UTF-8" > > > Pietro Di Primo wrote: >> Hi, I have a client server application, and want to tunnel it in this > way: >> a script select a free port in a random way, write the stunnel.conf >> file. >> with "accept=<the_found_port>" and "connect=<my_server>" and run my > client >> application passing to 127.0.0.1:<the_found_port> as server address. >> The problem is that i need to be sure that no other client or user can >> access <the_found_port> while the application is running. >> I guess that it should be possible to do it, but i wasn't able to. > > Simply bind the specific network interface with: > accept=127.0.0.1:<the_found_port> > > Best regards, > Mike > > > > ------------------------------ > > _______________________________________________ > stunnel-users mailing list > stunnel-users(a)mirt.net > http://stunnel.mirt.net/mailman/listinfo/stunnel-users > > > End of stunnel-users Digest, Vol 48, Issue 8 > ******************************************** > -- --- Ing. Pietro Di Primo INFN - Istituto Nazionale di Fisica Nucleare Grid Computing Research Area Via S. Sofia 78 95123 CATANIA (Italy) Voice : +39 095 378-5446/5517 Fax. : +39 095 378 5518 Mobile : +39 320 8591527 E-mail : pietro.diprimo(a)ct.infn.it Skype : pietrodiprimo

1 0

client-server tunnelling
by Pietro Di Primo 09 Jul '08

09 Jul '08

Hi, I have a client server application, and want to tunnel it in this way: a script select a free port in a random way, write the stunnel.conf file. with "accept=<the_found_port>" and "connect=<my_server>" and run my client application passing to 127.0.0.1:<the_found_port> as server address. The problem is that i need to be sure that no other client or user can access <the_found_port> while the application is running. I guess that it should be possible to do it, but i wasn't able to. Can anybody help me? Thanks -- --- Ing. Pietro Di Primo INFN - Istituto Nazionale di Fisica Nucleare Grid Computing Research Area Via S. Sofia 78 95123 CATANIA (Italy) Voice : +39 095 378-5446/5517 Fax. : +39 095 378 5518 Mobile : +39 320 8591527 E-mail : pietro.diprimo(a)ct.infn.it Skype : pietrodiprimo

1 0

FW: how to start ??
by dominique 07 Jul '08

07 Jul '08

Hello, First time I am using stunnel. 1. The issue I had is with Altiris and Exchange. All our Exchange Servers are SSL now and Altiris is not-SSL. Altiris is reading an exchange mailbox and gives back an error : "Error receiving data from socket". So the infrastructure is composed of two servers one Altiris Server not-SSL which is attempting to communicate to an Exchange Server with SSL. The Altiris Server is trying to read the email in one mailbox hosted by an Exchange SSL Server 2. Stunnel 4.25 x86-pc-mingw32-gnu with openssl 0.9.8h 28 May 2008 3. Windows 2003 Server environment 4. stunnel -d -r 5. 2008.07.07 13:34:41 LOG7[4748:1676]: RAND_status claims sufficient entropy for the PRNG 2008.07.07 13:34:41 LOG7[4748:1676]: PRNG seeded successfully 2008.07.07 13:34:41 LOG7[4748:1676]: Certificate: stunnel.pem 2008.07.07 13:34:41 LOG7[4748:1676]: Certificate loaded 2008.07.07 13:34:41 LOG7[4748:1676]: Key file: stunnel.pem 2008.07.07 13:34:41 LOG7[4748:1676]: Private key loaded 2008.07.07 13:34:41 LOG7[4748:1676]: SSL context initialized for service pop3 2008.07.07 13:34:41 LOG5[4748:1676]: stunnel 4.25 on x86-pc-mingw32-gnu with OpenSSL 0.9.8h 28 May 2008 2008.07.07 13:34:41 LOG5[4748:1676]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2008.07.07 13:34:41 LOG5[4748:3596]: No limit detected for the number of clients 2008.07.07 13:34:41 LOG7[4748:3596]: FD 284 in non-blocking mode 2008.07.07 13:34:41 LOG7[4748:3596]: SO_REUSEADDR option set on accept socket 2008.07.07 13:34:41 LOG3[4748:3596]: Error binding pop3 to 149.142.194.230:995 2008.07.07 13:34:41 LOG3[4748:3596]: bind: Can't assign requested address (WSAEADDRNOTAVAIL) (10049) 2008.07.07 13:34:41 LOG3[4748:3596]: Server is down 6. no file gcc. Thanks, Dominique Dominique_duchemin(a)hotmail.com

1 0

Using externally signed certificate with stunnel 4
by Tim Skirvin 07 Jul '08

07 Jul '08

I've got a comodo signed SSL certificate that I'm trying to use with stunnel4 to allow secure NNTP connections from a wide variety of clients. The certificate at least partially works; if I leave 'verify' off in the stunnel.conf file, then the service runs and users can connect, albeit while still having to verify the cert. But if I turn 'verify' on, then it doesn't work on *either* side. I've tried playing with CAfile and CApath without much luck. I'll attach my configuration files, the relevant pems, and some debugging information; is there something else I'm missing? I've already contacted comodo, and after several rounds of conversation they suggest I contact the list. Errors from the client side (note that I'm using a debug port here): + openssl s_client -connect news:565 -verify -debug verify depth is 0 CONNECTED(00000003) depth=2 /C=US/O=Entrust.net/OU=www.entrust.net/CPS incorp. by ref. (limits liab.)/OU=(c) 1999 Entrust.net Limited/CN=Entrust.net Secure Server Certification Authority verify error:num=19:self signed certificate in certificate chain verify return:0 6976:error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed:s3_clnt.c:951: ...and on the server, I get this: 2008.06.30 14:08:38 LOG7[10039:47679267941088]: nntps accepted FD=7 from 171.64.19.111:56122 2008.06.30 14:08:38 LOG7[10039:1073809760]: nntps started 2008.06.30 14:08:38 LOG7[10039:1073809760]: FD 7 in non-blocking mode 2008.06.30 14:08:38 LOG7[10039:1073809760]: FD 8 in non-blocking mode 2008.06.30 14:08:38 LOG7[10039:1073809760]: FD 9 in non-blocking mode 2008.06.30 14:08:38 LOG7[10039:47679267941088]: Cleaning up the signal pipe 2008.06.30 14:08:38 LOG6[10039:47679267941088]: Child process 10247 finished with code 0 2008.06.30 14:08:38 LOG7[10039:1073809760]: Connection from 171.64.19.111:56122 permitted by libwrap 2008.06.30 14:08:38 LOG5[10039:1073809760]: nntps connected from 171.64.19.111:56122 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): before/accept initialization 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): SSLv3 read client hello A 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): SSLv3 write server hello A 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): SSLv3 write certificate A 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): SSLv3 write certificate request A 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL state (accept): SSLv3 flush data 2008.06.30 14:08:38 LOG7[10039:1073809760]: SSL alert (read): fatal: unknown CA 2008.06.30 14:08:38 LOG3[10039:1073809760]: SSL_accept: 14094418: error:14094418:SSL routines:SSL3_READ_BYTES:tlsv1 alert unknown ca 2008.06.30 14:08:38 LOG5[10039:1073809760]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.06.30 14:08:38 LOG7[10039:1073809760]: nntps finished (0 left) Basic Requested Information that wasn't supplied above: * stunnel 4.18-2 on Debian etch (2.6.18-6-686 #1 SMP) * Running standalone with '/usr/bin/stunnel4 /etc/news/stunnel.conf' * libc6, no gcc, OpenSSL 0.9.8c-4etch3 * The log on startup: 2008.06.30 14:00:15 LOG7[26276:3083523776]: Snagged 64 random bytes from /root/.rnd 2008.06.30 14:00:15 LOG7[26276:3083523776]: Wrote 1024 new random bytes to /root/.rnd 2008.06.30 14:00:15 LOG7[26276:3083523776]: RAND_status claims sufficient entropy for the PRNG 2008.06.30 14:00:15 LOG7[26276:3083523776]: PRNG seeded successfully 2008.06.30 14:00:15 LOG7[26276:3083523776]: Configuration SSL options: 0x01000000 2008.06.30 14:00:15 LOG7[26276:3083523776]: SSL options set: 0x01000000 2008.06.30 14:00:15 LOG7[26276:3083523776]: Certificate: /etc/ssl/certs/news-stunnel.pem 2008.06.30 14:00:15 LOG7[26276:3083523776]: Certificate loaded 2008.06.30 14:00:15 LOG7[26276:3083523776]: Key file: /etc/ssl/private/news-stunnel.key 2008.06.30 14:00:15 LOG7[26276:3083523776]: Private key loaded 2008.06.30 14:00:15 LOG7[26276:3083523776]: Loaded verify certificates from /etc/ssl/certs/comodo.cert 2008.06.30 14:00:15 LOG7[26276:3083523776]: SSL context initialized for service nntps - Tim Skirvin (tskirvin(a)stanford.edu) -- Information Technology Services http://www.stanford.edu/~tskirvin/ System Software Developer, Unix Team Stanford University

1 1

Windows setup
by Koenraad Lelong 04 Jul '08

04 Jul '08

Hello, I would like to know if it is possible to run a silent setup of the Windows-setup of stunnel. I'm including it in my own setup of my own application so I would like to avoid the interaction of the stunnel-setup. Second question is if it is bad to "install" stunnel more than once on a computer. Until now I had a setup for new installations and an update for updates of my application. This seems not to work because some users use the setup to update an existing installation and with this they run the install of stunnel for the second time. If it is bad, how do I detect the existance of stunnel, and what version it is, so I could update stunnel if it's outdated. Regards, Koenraad Lelong. P.S. If this is the wrong list for such questions I apologize. I would like to know the right list.

1 0

Re: [stunnel-users] Stunnel 4.21 99% cpu ultilization
by Fat Wallet 03 Jul '08

03 Jul '08

Hi, If anyone have a clue on what I should do to fix this issue, I would really appreciated: One of the stunnel process is running at 99% and in R state and the time just keep incrementing. [root@root]# ps auxw|grep stunnel root 781 0.0 0.0 6520 2684 ? T 11:01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 2853 0.0 0.0 3940 604 ttyp2 S+ 11:37 0:00 grep stunnel root 16554 0.0 0.0 3992 788 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16555 0.0 0.0 3992 788 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16556 0.0 0.0 3992 788 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16557 0.0 0.0 3992 788 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16558 0.0 0.0 3992 788 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16561 0.0 0.0 6520 2684 ? Ts Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 16656 0.0 0.0 6520 2684 ? S Jul01 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 23954 0.0 0.0 6520 2684 ? S 08:00 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 23955 99.0 0.0 6520 2684 ? R 08:00 189:52 /usr/sbin/stunnel /etc/stunnel/stunnel.conf root 31521 0.0 0.0 6520 2684 ? S 10:30 0:00 /usr/sbin/stunnel /etc/stunnel/stunnel.conf Thanks, Mike ----- Original Message ---- From: Aron Griffis <agriffis(a)n01se.net> To: Fat Wallet <fat8wallet(a)yahoo.com> Cc: stunnel-users(a)mirt.net Sent: Monday, June 30, 2008 11:04:44 AM Subject: Re: [stunnel-users] Stunnel 4.21 99% cpu ultilization Fat Wallet wrote: [Mon Jun 30 2008, 01:28:11PM EDT] > Can someone please help with this issue? This 100% cpu > utilization issue happens randomly on multiprocess cpu. strace, ltrace, build stunnel with debugging symbols and attach gdb to the spinning process, etc. Is your name really Fat Wallet?

2 1

protect the private key, while running stunnel in server mode as a windows service
by sandeep siroya 03 Jul '08

03 Jul '08

Hi, I have stunnel in server mode running as a windows service on a Win2003 Server system. I need to give this system to a different company for testing their client against our server. How do I protect my private key on stunnel, while letting the 3rd company being able to login as administrator on stunnel's win2003 system and start/stop stunnel service? I know one can protect the private key with a passphrase in stunnel, but that does not serve my purpose as stunnel in this case should be able to start/stop without any user-intervention and I cannot store the passphrase in cleartext. thanks , ss

1 0

Peer suddenly disconnected
by victor.harutyunyan＠arca.am 02 Jul '08

02 Jul '08

Hello there, I have installed stunnel server on FreeBSD. I configure it like this cert=server_cert.pem key=private\server_key.pem socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 debug = 7 verify=2 CAfile=test_cert.pem [ssl_host_proxy] accept=172.24.150.15:1234 connect=192.168.1.1:10500 TIMEOUTidle=120 Client side is not a computer it's small device. I generate client certificates in this manor: echo off del test_ck.pem /q # delete previous test.ck.pem file echo Starting New test_ck file Generation C:\OpenSSL\bin\openssl.exe req -new -x509 -newkey rsa:2048 -sha1 -out test_cert.pem -keyout test_key.pem -nodes -config sample_ck.cnf -days 1825 -batch C:\OpenSSL\bin\openssl.exe pkcs8 -in test_key.pem -topk8 -out test_pk8.pem -passout pass:"TestPassword0001" -noiter -v1 PBE-SHA1-3DES del test_key.pem /q copy test_cert.pem+test_pk8.pem test_ck.pem /y del test_pk8.pem /q Echo New test_ck.pem file has the following sections: findstr "BEGIN END" test_ck.pem pause test_cert.pem is on a stunnel server as CAfile and test_ck.pem is on my small device which is capable to connect via stunnel. And I generate other self-signed CA, and generate one certificate request and sign it via this CA, and install that key and certificate on server. cert=server_cert.pem key=private\server_key.pem All is working fine, but after one year my server certificate has expired, when I regenerate new server certificate and try to connect, I see in server log this: Child process 6094 finished with code 0 SSL state (accept): before accept initialization SSL state (accept): SSLv3 read client hello A SSL state (accept): SSLv3 write server hello A SSL state (accept): SSLv3 write certificate A SSL state (accept): SSLv3 write server done A SSL state (accept): SSLv3 flush data SSL_accept: Peer suddenly disconnected And my device can't connect to the server, I haven't any client side logs and error messages, only NOT DELIVERED in my device. :( Can you please help me to solve this problem, I am trying many many times, I was very angry.. Thank you very much. Victor Harutyunyan. P.S. Sorry for my bad english.

1 0

Peer suddenly disconnected
by victor.harutyunyan＠arca.am 02 Jul '08

02 Jul '08

Hello there, I have installed stunnel server on FreeBSD. I configure it like this cert=server_cert.pem key=private\server_key.pem socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 debug = 7 verify=2 CAfile=test_cert.pem [ssl_host_proxy] accept=172.24.150.15:1234 connect=192.168.1.1:10500 TIMEOUTidle=120 Client side is not a computer it's small device. I generate client certificates in this manor: echo off del test_ck.pem /q # delete previous test.ck.pem file echo Starting New test_ck file Generation C:\OpenSSL\bin\openssl.exe req -new -x509 -newkey rsa:2048 -sha1 -out test_cert.pem -keyout test_key.pem -nodes -config sample_ck.cnf -days 1825 -batch C:\OpenSSL\bin\openssl.exe pkcs8 -in test_key.pem -topk8 -out test_pk8.pem -passout pass:"TestPassword0001" -noiter -v1 PBE-SHA1-3DES del test_key.pem /q copy test_cert.pem+test_pk8.pem test_ck.pem /y del test_pk8.pem /q Echo New test_ck.pem file has the following sections: findstr "BEGIN END" test_ck.pem pause test_cert.pem is on a stunnel server as CAfile and test_ck.pem is on my small device which is capable to connect via stunnel. And I generate other self-signed CA, and generate one certificate request and sign it via this CA, and install that key and certificate on server. cert=server_cert.pem key=private\server_key.pem All is working fine, but after one year my server certificate has expired, when I regenerate new server certificate and try to connect, I see in server log this: Child process 6094 finished with code 0 SSL state (accept): before accept initialization SSL state (accept): SSLv3 read client hello A SSL state (accept): SSLv3 write server hello A SSL state (accept): SSLv3 write certificate A SSL state (accept): SSLv3 write server done A SSL state (accept): SSLv3 flush data SSL_accept: Peer suddenly disconnected And my device can't connect to the server, I haven't any client side logs and error messages, only NOT DELIVERED in my device. :( Can you please help me to solve this problem, I am trying many many times, I was very angry.. Thank you very much. Victor Harutyunyan. P.S. Sorry for my bad english.

1 0

socket not getting closed
by Manjunath S Karikatti 02 Jul '08

02 Jul '08

Hello, We are using stunnel to wrap the tcp connetion between the our client session used by users and the server process. Due to some reason when client session terminates abruptly (either due to user's PC crash or network outage) the stunnel process does not terminate. We see that the socket on client PC is closed but stunnel process is still alive with socket for that PC in established state. This is on AIX platform. Do not know this is issue with stunnel process or something on AIX needs to be tuned (like the tcp_keepidle value). Thanks for any info and help. Thanks and Regards, Manjunath

1 0

Re: [stunnel-users] Stunnel 4.21 99% cpu ultilization
by Fat Wallet 01 Jul '08

01 Jul '08

Thanks for your reply. I was hoping to see if someone have a similar issue since there was an issue with stunnel 4.20 version. Thanks, Mike ----- Original Message ---- From: Aron Griffis <agriffis(a)n01se.net> To: Fat Wallet <fat8wallet(a)yahoo.com> Cc: stunnel-users(a)mirt.net Sent: Monday, June 30, 2008 11:04:44 AM Subject: Re: [stunnel-users] Stunnel 4.21 99% cpu ultilization Fat Wallet wrote: [Mon Jun 30 2008, 01:28:11PM EDT] > Can someone please help with this issue? This 100% cpu > utilization issue happens randomly on multiprocess cpu. strace, ltrace, build stunnel with debugging symbols and attach gdb to the spinning process, etc. Is your name really Fat Wallet?

1 0

Stunnel v4.25 vs v4.20
by James Moe 30 Jun '08

30 Jun '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello, I upgraded to v4.25 and stunnel aborts right away. v4.20 works just fine. What is protocol family 47? ====[ log file ]==== 2008.06.26 11:31:01 LOG5[282:8553]: stunnel 4.25 on i386-pc-os2-emx with OpenSSL 0.9.8h 28 May 2008 2008.06.26 11:31:01 LOG5[282:8553]: Threading: SSL:ENGINE Sockets:SELECT,IPv4 2008.06.26 11:31:01 LOG3[282:8553]: Error binding pop3.sma.com_L110-R995 to 127.0.0.1:110 2008.06.26 11:31:01 LOG3[282:8553]: bind: Address family not supported by protocol family (47) ====[ conf file ]==== socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 client = yes output = G:/c/voice/pmmdev/testcase/bin/stunnel.log verify = 0 debug = 5 cert = g:/c/voice/pmmdev/testcase/bin/sma-test.pem [pop3.sma.com_L110-R995] accept = localhost:110 connect = pop3.sma.com:995 ;# protocol = pop3 [smtpauth.earthlink.net_L6025-R587] accept = localhost:6025 connect = smtpauth.earthlink.net:587 protocol = smtp [sohnen-moe.com:L6125:R587] accept = localhost:6125 connect = sohnen-moe.com:587 protocol = smtp [smtp.sma.com_L25-R587] accept = localhost:25 connect = smtp.sma.com:587 protocol = smtp - -- jimoe (at) sohnen-moe (dot) com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (OS/2) iD8DBQFIY+DkzTcr8Prq0ZMRArW+AJ4uOJFOGyluIj52DG4cM62jMPs8NACfbug2 oVsLAXJ+fH1Raj568dGn2mM= =JSCM -----END PGP SIGNATURE-----

1 1

Re: [stunnel-users] Stunnel 4.21 99% cpu ultilization
by Fat Wallet 30 Jun '08

30 Jun '08

Hi, Can someone please help with this issue? This 100% cpu utilization issue happens randomly on multiprocess cpu. Thanks! ----- Original Message ---- From: Fat Wallet <fat8wallet(a)yahoo.com> To: stunnel-users(a)mirt.net Sent: Monday, June 23, 2008 10:39:16 AM Subject: Stunnel 4.21 99% cpu ultilization Hi All, I've been googling for this issue but it seems like google is not my friend at the moment. I am running Linux with 4 cpus and top command shows stunnel at 99-100% cpu. Please help. //stunnel configuration file # PID is created inside chroot jail pid = /var/run/stunnel.pid # debug debug = 7 output = /var/log/stunnel.log # Authentication stuff verify = 2 cert = /etc/stunnel/cert.pem CAfile = /etc/stunnel/ca.pem #Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 # Use it for client mode client = yes /////////// stunnel version stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.7a Feb 19 2003 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /var/run/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none //openssl version OpenSSL 0.9.7a Feb 19 2003 //Redhat version Red Hat Linux release 9 (Shrike) root]#cat /proc/cpuinfo | grep processor | wc -l 4 http://stunnel.mirt.net/mailman/listinfo/stunnel-users

2 1

UPNP & Nat-PMP with Stunnel
by Dean Shavit 26 Jun '08

26 Jun '08

Stunnel works great for me, but with dynamic port forwarding, I've not been able to get it working. I suspect it's because stunnel is a bit confusing about which port to reply to... Anyone else found a way to crack this nut?

1 0

How do I kill stunnel?
by Grant Edwards 25 Jun '08

25 Jun '08

I've got an application that starts stunnel using equivalent of a fork/exec. The stunnel process then appears to fork off multiple copies of itself. Before the application exits, it needs to stop stunnel. I send a SIGINT to the PID that was created using the fork, but only the first "master" copy terminates -- the master stunnel process doesn't terminate its children when it exits. So I end up with a bunch of "orphaned" stunnel processes. They're not zombies, they're process that are either in state R or S (runnable or sleeping). How do I tell stunnel to not only terminate, but to terminate whatever child processes it has created? -- Grant Edwards grante Yow! Zippy's brain cells at are straining to bridge visi.com synapses ...

2 4

Stunnel config files incompatible between 4.05 and 4.21 ?
by Brian McKee 25 Jun '08

25 Jun '08

Hi All, I have a stunnel config file that works fine on an older Mandrake machine running stunnel 4.05. If I use that same config file on a new Ubuntu machine running 4.21 it just dies with "/etc/stunnel/stunnel.pem: No such file or directory" IIUC, running in client mode it doesn't need a .pem file, and I didn't create any certifcates on the Mandrake machines. Do I need to generate a cert for some reason? I've attached the config below. Brian =============================== client = yes [telnet] accept = 23 connect = some.ip.number.removed:992 [as-central] accept = 8470 connect = some.ip.number.removed:9470 [as-database] accept = 8471 connect = some.ip.number.removed:9471 [as-rmtcmd] accept = 8475 connect = some.ip.number.removed:9475

1 0

Is the "HTTP Proxy connect" feature implemented?
by alex Wright 24 Jun '08

24 Jun '08

Hey, I'm trying to use stunnel in the following scenerio: ssh-client -> stunnel (to ssl) -> Proxy -> server(ssl to ssh). I downloaded stunnel 4.25, but I hadn't noiticed any support for this. However, in the "patches" page (http://www.stunnel.org/patches/) - three patches seem to be suitable: (connect-proxy_dunbar.patch<http://www.stunnel.org/patches/patches/connect-proxy_dunbar.patch>, connect-proxy.mwald.patch<http://www.stunnel.org/patches/patches/connect-proxy.mwald.patch>) but are for version 4.05 (whichwill probably be a pain to merge with current version). When I stumbled upon the Changelog ( http://www.stunnel.org/download/ChangeLog.txt) I noticed that version 4.15, has support for "Client mode CONNECT protocol" - which is what I need - and altough i had looked hard, I haven't found any documentation to it. So, is the above feature what I want (and if so - how do I use it?) or do I need to update the patches to the current version? Thanks in advance, Alex.

1 0

Stunnel 4.21 99% cpu ultilization
by Fat Wallet 23 Jun '08

23 Jun '08

Hi All, I've been googling for this issue but it seems like google is not my friend at the moment. I am running Linux with 4 cpus and top command shows stunnel at 99-100% cpu. Please help. //stunnel configuration file # PID is created inside chroot jail pid = /var/run/stunnel.pid # debug debug = 7 output = /var/log/stunnel.log # Authentication stuff verify = 2 cert = /etc/stunnel/cert.pem CAfile = /etc/stunnel/ca.pem #Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 # Use it for client mode client = yes /////////// stunnel version stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.7a Feb 19 2003 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /var/run/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none //openssl version OpenSSL 0.9.7a Feb 19 2003 //Redhat version Red Hat Linux release 9 (Shrike) root]#cat /proc/cpuinfo | grep processor | wc -l 4 http://stunnel.mirt.net/mailman/listinfo/stunnel-users

1 0

Using Stunnel on Vista
by John Lallemand 18 Jun '08

18 Jun '08

We have a system that needs a TN3270 client and Stunnel to connect on XP. Currently, OnNet Host/Rumba is the most popular TN3270 client in use. But when we try these programs on Vista, the connection times out. 1) Is Stunnel Vista compatible? 2) If it is, does Stunnel need specific settings to work on Vista? Thank you! ------------------------------ John Lallemand Software Release Manager University of California, Santa Cruz

2 1

Problem with multiple stunnel jobs
by Carter Browne 17 Jun '08

17 Jun '08

A couple of days ago, I upgrade from 4.15 to 4.23 on a Linux system running Redhat 8. At the same time I went from openssl 0.9.8g to 0.9.8h. No changes were made to the configuration file. After the change, rather than starting a single process, it starts 6 processes. The process number of the lowest number process is the one that is put into the /var/run/stunnel.pid file. When I try to stop stunnel, using /etc/init.d/stunnel stop, the lowest number process is kill, and the highest number process seems to be running continuously - although it is not a hard cpu loop, but it keeps on accumulating run time. The number of processes may not be a new phenomenon, but the inability to kill itself is. I normally restart the stunnel on a daily basis to keep the log files manageable, but with this new behavior, the /etc/init.d/strunnel restart command fails because the ports are still in use. Any ideas on how to fix this problem. Thanks, Carter -- Carter Browne CBCS cbrowne(a)cbcs-usa.com 781-721-2890

2 1

Goldmine 8 & Advisormail SSL problems
by Stephen E. Rischall 17 Jun '08

17 Jun '08

1 0

stunnel service problems...
by Jeremy Paje 10 Jun '08

10 Jun '08

Has anyone had any problems with the windows stunnel service not starting automatically? We install stunnel as a service (stunnel.exe -install) We've had it previously installed on 600+ machines with no problems, but recently (noticed on slower machines running XP) the stunnel service does not start...but it happens randomly only on slower machines. The following errors are in the event log... Timeout (30000 milliseconds) waiting for the stunnel service to connect. (Event ID ID 7009) AND The stunnel service failed to start due to the following error: The service did not respond to the start or control request in a timely fashion. (Event ID 7000) I've tried setting the service to restart if failed, and to delay 2 minutes before restart, etc with no luck. If you start the stunnel manually all is well, it just doesn't start on bootup. Any suggestions? Thanks JP

1 0

Re: [stunnel-users] stunnel compatibility question
by Brian Hatch 09 Jun '08

09 Jun '08

Roughly around 2008-06-09 11:34 -0400, Patel Dippen-CDP054 kibitzed: > In this scenario, can Stunnel listen on the UDP port on the local No. Stunnel will not listen on UDP ports. Stunnel will not send UDP packets. Stunnel doesn't support UDP. -- Brian Hatch The problem is that you'd Systems and probably make something Security Engineer useful of my life that http://www.ifokr.org/bri/ I would never do. Every message PGP signed

2 1

Fw: Re: stunnel compatibility question
by Jeffrey Buck 06 Jun '08

06 Jun '08

1 0

Re: [stunnel-users] stunnel compatibility question
by Brian Hatch 05 Jun '08

05 Jun '08

At almost 2008-06-05 09:37 -0400, Patel Dippen-CDP054 cajoled: > I guess what I am stating is can Stunnel listen on a UDP port The answer to that question is 'no, it cannot.' -- Brian Hatch "I was looking at bringing Systems and up some new servers on Security Engineer imaginary hardware." http://www.ifokr.org/bri/ --Jason Every message PGP signed

1 0

stunnel compatibility question
by Jeffrey Buck 04 Jun '08

04 Jun '08

2 4

Re: [stunnel-users] stunnel compatibility question
by Brian Hatch 03 Jun '08

03 Jun '08

In the neigborhood of 2008-06-03 07:58 -0400, Patel Dippen-CDP054 argued: > Will Stunnel work for UDP data? Stunnel only accepts TCP connections. > We have traffic going into the Server on UDP and outbound from the Server on TCP. You'll want an actual VPN protocol to handle that then. Or consider using a better protocol - what you describe makes the firewalls cry. > My understanding is that OpenSSL implemented the DTLS protocol (TFC > 4347: to support UDP data) in version 0.9.8 (OpenSSL: a free and popular > implementation of TLS added support for DTLS with version 0.9.8.) Not supported in Stunnel currently. -- Brian Hatch Always yield to temptation, Systems and because it may not pass Security Engineer your way again. http://www.ifokr.org/bri/ Every message PGP signed

1 0

Is anyone using Stunnel for tunnelling Voip?
by Errol Samuels 03 Jun '08

03 Jun '08

Hello All, I am currently researching a solution for tunnelling VoIP traffic since VoIP is blocked in the country concerned. It appears that the ISP is targeting OpenVPN and OpenSwan and blocking them so I want to try something different. My solution based on OpenVPN worked great until it got blocked by the ISP about a week ago. It was using an OpenVPN server hosted in a US Datacenter and the VPN clients were simply routers with third party Linux firmware with the OpenVPN client embedded in the firmware. I am also investigating the possibility of tunnelling OpenVPN through SSH but I have to watch my IP overheads here because the ISP watches the upload bandwidth as well! Has anyone attempted or had any success using stunnel for a similar scenario as I have mentioned? Thanks in Advance Regards, Errol

3 9

Can I put an accept+connect pair anywhere in the conf file?
by Ben Stover 02 Jun '08

02 Jun '08

The *.conf file is divided into several sections: [imaps] .... [pop3s] .... [smtps] ..... Are these "section headers" just comments and can I put my accept+connect pair on an arbitrary place in the *.conf file (as long as they are preceeded by their own section header [foobar] ? Ben

2 1

Connecting to mySQL from Windows
by Alistair Scott 02 Jun '08

02 Jun '08

Hi everyone I have a mySQL database that I need to connect to through SSH. I have successfully setup a connection to it using Plink. However, I'm running this from a batch file and I'd like a more elegant solution - that's where stunnel comes in I think. Rather than running a batch file, I'd like to run the connection as a Windows service - which I think this is achievable with stunnel. Trouble is, I can't work out how to do it. This is what I'm trying to achieve... Windows computer connects to mySQL server through SSH. Forwards a port to the mySQL server port 3306. Application on Windows computer connects to local forwarded port & therefore connects to the mySQL db at the remote end. I was hoping that someone could give me a pointer as to how I can set this up. Thanks in advance. Al

1 0

stunnel 4.25 released
by Michal Trojnara 01 Jun '08

01 Jun '08

Dear Users, Version 4.25, 2008.06.01, urgency: MEDIUM: * New features - Win32 DLLs for OpenSSL 0.9.8h. * Bugfixes - Spawning libwrap processes delayed until privileges are dropped. - Compilation fix for systems without struct msghdr.msg_control. Home page/download: http://stunnel.mirt.net/ sha1sum for stunnel-4.24.tar.gz file: fc6d61fad996f750c76ea627c5dd9f789af0eaf6 Best regards, Mike

1 0

Re: [stunnel-users] stunnel compatibility question
by Brian Hatch 30 May '08

30 May '08

Right around 2008-05-30 13:51 -0400, Patel Dippen-CDP054 suggested: > Is the true? I have a similar issue. I have never implemented SSL or > Stunnel. But, on whatever readup I have done so far, it seems Stunnel is > a tunneling protocol (IPnIP). So, you do have to have Stunnel on the > other end to extract the IP packet. Stunnel puts data inside SSL, inside TCP. Just like HTTPS is HTTP (data) inside SSL, inside TCP. You can think of the cleartext data inside the SSL tunnel for both of those examples. Yes, the data needs to be extracted from the SSL stream in both cases, but the application that gets it doesn't need to know diddly about it. > I mean if you are saying it will work, that's great news for me. It works if Stunnel is on one side and an SSL client or server is on the other side. > Our Server will have OpenSSL implementation but we plan to deploy > Stunnel on the Client machine. That's what Stunnel is for. -- Brian Hatch "Sorry, sorry, Systems and sorry, much apologizings." Security Engineer http://www.ifokr.org/bri/ Every message PGP signed

1 0

Need class name for stunnel
by Bill Murphy 28 May '08

28 May '08

I'm using stunnel to send and receive emails in a Microsoft Access 2007 application. I would like to be able to programmatically close stunnel when the user exits a form in the application which is used for sending and receiving emails. I can close other programs running in Windows with the code in this api: http://www.mvps.org/access/api/api0025.htm This code requires the "class name" of the application while it is running in Windows in order to close it. Stunnel.exe does not appear to have a class name. I've used the following additional api to display the class names of all currently running apps: http://www.mvps.org/access/api/api0013.htm Stunnel.exe does not show up in the list of running apps produced by this code. Is there another way to determine the stunnel class name? Bill

1 0

Re: [stunnel-users] Problem with Stunnel and OpenBSD
by Juan J. Martínez 23 May '08

23 May '08

Hello, I'm debugging with openssl s_client... >From inside the box: $ openssl s_client -connect blackshell.usebox.net:995 CONNECTED(00000004) depth=0 /C=ES/ST=Valencia/L=Valencia/O=usebox.net/OU=blackshell/CN=blackshell.usebox.net verify error:num=18:self signed certificate verify return:1 depth=0 /C=ES/ST=Valencia/L=Valencia/O=usebox.net/OU=blackshell/CN=blackshell.usebox.net verify return:1 ... And works fine. >From outside: $ openssl s_client -connect blackshell.usebox.net:995 CONNECTED(00000003) (and halts for n minutes) I hope this helps, even I don't know the difference between "CONNECTED(00...4)" and "CONNECTED(00...3)". Regards, Juanjo -- desarrollo y sistemas: http://www.usebox.net/ página personal: http://www.usebox.net/jjm/

1 1

STUNNEL: Thunderbird with Exchange and only inbound port 443
by Fatma 23 May '08

23 May '08

Hello All, A quick question regarding the stunnel ability to work with the following scenario. Scenario is: I have an Exchange Server sitting on the external perimeter with Outlook Web Access (OWA) enabled and it is the frontend and backend mail server as well. And I have a remote client which has a Thunderbird Client (for example) that I need it to connect to the Exchange Server to retrieve emails. The only inbound port to the Exchange Server from a network perspective is HTTPS-443 and no other ports are open on the firewall. Now the question is how can a client thunderbird connect to the Exchange Server/OWA with only 443 port open. So what i was thinking is and tell me if this works? If i install two instances of STUNNEL one on the remote client and one on the exchange server and the following happens. on the client the stunnel listens on port 25 SMTP and encrypts and forwards to port 443 then the server recieves and listens on port 443 and decrypts and forwards to port 110 POP3 will that work actually? hmmm i wish i explained that pretty well. if not then how can it work with the given scenario. cheerz.

1 0

Re: [stunnel-users] Wrapping a mail client - multiple servers
by Ludolf Holzheid 21 May '08

21 May '08

On Wed, 2008-05-21 14:39:42 +0100, Daniel James wrote: > [..] > > Can I, for example, put multiple [SSMTP] sections and multiple [POP3S] > sections in one stunnel.conf? I don't know. But if you can, you don't want to. The section headers don't specify a protocol. They simply give the section a name (for log output, e.g.). There is no problem in naming sections [first-POP3S] and [second-POP3S]. HTH, Ludolf -- --------------------------------------------------------------- Ludolf Holzheid Tel: +49 621 339960 Bihl+Wiedemann GmbH Fax: +49 621 3392239 Floßwörthstraße 41 e-mail: lholzheid(a)bihl-wiedemann.de D-68199 Mannheim, Germany ---------------------------------------------------------------

1 0

Wrapping a mail client - multiple servers
by Daniel James 21 May '08

21 May '08

I use a mail client that does not directly support SSL. I can use stunnel to provide the security layer when I connect to a SSMTP and/or POP3S server (under windows). This works fine, whether I run stunnel as a background application or as a Windows service. I'd like, however, to be able to connect to multiple mail servers over SSL. I'm not sure how I might achieve this? What would be nice would be get stunnel to forward POP3/SMTP communication on different local ports to POP3S/SSMTP on different remote servers. Can I do this? It appears to be possible to run multiple copies of stunnel as applications, and to get each copy to take its configuration from a different stunnel.conf file, so that different copies of stunnel could listen on different local ports and each could proxy the traffic it receives to a different remote server (I haven't tried this -- and it's not an ideal solution -- but it looks as though it should work). It does not appear to be possible, however, to create two instances of the stunnel service running with different configuration files. Is it possible to set up multiple listening ports proxying to different remote servers in a single configuration file? If so, what's the syntax for that? Cheers, Daniel.

1 0

Stunnel 4 Multiple Processes on Ubuntu 8.04
by Armen Ayvazian 19 May '08

19 May '08

Greetings. I'm running the stock stunnel 4.21 on Ubuntu 8.04 (version and configs listed below) as a stand-alone daemon to pass connections between ssl-enabled SMTP, POP3, and IMAP clients and their respective servers, which run on the same machine as stunnel. In previous instances of Stunnel (v4.15) I've managed to successfully get the daemon to switch to a non-root user and use only a single process. I now find, however, that I get 6 processes and 5 continue to run as root with only one (albeit the one listening on the specified sockets) switching down to the specified user. Does anyone have any idea why I both can't set uid/gid and why I've now got so many processes? # ps aux | grep stunnel root 22702 0.0 0.1 3604 428 pts/1 S 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf root 22703 0.0 0.1 3604 428 pts/1 S 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf root 22704 0.0 0.1 3604 428 pts/1 S 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf root 22705 0.0 0.1 3604 428 pts/1 S 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf root 22706 0.0 0.1 3604 428 pts/1 S 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf stunnel4 22707 0.0 0.2 3720 932 ? Ss 11:24 0:00 /usr/bin/stunnel4 /etc/stunnel/stunnel.conf # lsof -i | grep stunnel stunnel4 22707 stunnel4 11u IPv4 55934 TCP emily:pop3s (LISTEN) stunnel4 22707 stunnel4 12u IPv4 55935 TCP emily:imaps (LISTEN) stunnel4 22707 stunnel4 13u IPv4 55936 TCP emily:ssmtp (LISTEN) I'm invoking stunnel simply as /usr/sbin/stunnel4 /etc/stunnel4/stunnel.conf: cert = /etc/stunnel/stunnel.pem sslVersion = SSLv3 chroot = /var/lib/stunnel4/ setuid = stunnel4 setgid = stunnel4 pid = /stunnel4.pid socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 [pop3s] accept = 64.5.53.172:995 connect = 64.5.53.172:110 [imaps] accept = 64.5.53.172:993 connect = 64.5.53.172:143 [ssmtp] accept = 64.5.53.172:465 connect = 64.5.53.172:25 Version Info: stunnel 4.21 on i486-pc-linux-gnu with OpenSSL 0.9.8g 19 Oct 2007 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /var/run/stunnel4.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = AES:ALL:!aNULL:!eNULL:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none

1 0

stunnel and OCSP verification: strange behaviour
by Andreas Ntaflos 19 May '08

19 May '08

Hi list, [My apologies, I accidentally tried to send this message to stunnel-announce earlier.] I have a quick question regarding the use of stunnel with verification against an OCSP responder. I am using stunnel 4.20 and (for testing) 4.23 but I don't see any difference in the behaviour of both. The main problem is that stunnel accepts and establishes connections from clients even if those clients had their certificates revoked and the OCSP responder correctly informs stunnel about it. Consider the following configuration: foreground = yes CAfile = /path/to/cacert.pem cert = /path/to/stunnel_cert.pem key = /path/to/stunnel_key.pem debug = 7 [clientauth] accept = localhost:43434 connect = localhost:43433 verify = 2 OCSP = http://localhost:43435 The OCSP responder runs on localhost:43435 as you can see. Client verification is enabled. Using openssl to manually check against the responder works fine: openssl ocsp -issuer cacert.pem -CAfile cacert.pem -url \ http://localhost:43435 -cert revoked_cert.pem Response verify OK /path/to/revoked_cert.pem: revoked This Update: May 15 15:38:46 2008 GMT Next Update: May 15 17:04:32 2008 GMT Revocation Time: May 15 15:38:26 2008 GMT The relevant portion of stunnel's debug output looks like this: [...] Starting OCSP verification FD 8 in non-blocking mode connect_wait: waiting 10 seconds connect_wait: connected OCSP server connected FD 8 in blocking mode FD 8 in non-blocking mode OCSP response received OCSP verification passed: status=0, reason=-1208427072 VERIFY OK: depth=1, /CN=The CA/C=AT/ST=SomeState/L=SomeCity/O=The \ Organisation/emailAddress=camaster(a)the-organisation.invalid Starting OCSP verification FD 8 in non-blocking mode connect_wait: waiting 10 seconds connect_wait: connected OCSP server connected FD 8 in blocking mode FD 8 in non-blocking mode OCSP response received OCSP verification passed: status=1, reason=-1 VERIFY OK: depth=0, /C=AT/ST=SomeState/O=The Organisation/CN=this is a \ revoked cert SSL state (accept): SSLv3 read client certificate A [...] As far as I can see the OCSP verification works fine here, too, and at depth 0 stunnel finds the certificate to have been revoked (status=1 indicates that, I believe). The connection, however, is not rejected. Is there any way to have stunnel reject the connection when the client certificate has been revoked? Where would I configure this? The only way I have found yet was to patch src/verify.c but surely there must be a more convenient approach to this? On a related matter: if stunnel cannot and will not reject the connection upon a negative response from the OCSP, what would be the best way to script stunnel so that it rejects such a connection? I am asking because I am in the process of writing a few Python scripts that incorporate stunnel to aid in testing of some SSL-speaking components. Such a client component without a valid, non-revoked certificate must not be able to connect. I mean what good is OCSP verification if the result of the process, i.e. the responders answer "the certificate is valid" or "the certificate is no longer valid" doesn't matter to stunnel? Am I missing the point somewhere? I'd appreciate any insights on this. Kind regards, Andreas -- Andreas "daff" Ntaflos Vienna, Austria GPG Fingerprint: 6234 2E8E 5C81 C6CB E5EC 7E65 397C E2A8 090C A9B4

2 2

Reg:IPV6 support on stunnel
by Sandhya 15 May '08

15 May '08

> Stunnel version: > stunnel 4.09 on armv5tejl-unknown-linux-gnu PTHREAD+POLL+IPv6+LIBWRAP with > OpenSSL 0.9.7e 25 Oct 2004 > > My conf file Information: > cert = /usr/local/www/certs/server.pem > key = /usr/local/www/certs/privkey-server.pem > pid = /var/run/stunnel_kvm.pid > client = no > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > socket = a:TCP_NODELAY=1 > [ms] > accept = 1234 > connect = 1236 > > Please find the Config.log file attached. > > Please let me know if any other information is needed or you want me try > some commands on stunnel. > > Thanks for the help > > Sandhya R > > ----- Original Message ----- > From: "Michal Trojnara" <Michal.Trojnara(a)mobi-com.net> > To: <stunnel-users(a)mirt.net> > Sent: Wednesday, May 14, 2008 7:06 PM > Subject: Re: [stunnel-users] Reg:IPV6 support on stunnel > > >> >> Sandhya wrote: >> >>> I have enabled IPV6 ip When my client tries to connect to my ipv6 ip and >>> the port(stunnel listens) where my it is listening...Its failed to >>> connect. >> >> Please send us: >> - the output of "stunnel -version", >> - your stunnel.conf file, >> - your debug level logs. >> >> Best regards, >> Mike >> >> _______________________________________________ >> stunnel-users mailing list >> stunnel-users(a)mirt.net >> http://stunnel.mirt.net/mailman/listinfo/stunnel-users >> >

1 1

Fw: Reg:IPV6 support on stunnel
by Sandhya 14 May '08

14 May '08

----- Original Message ----- From: Sandhya To: stunnel-users(a)mirt.net Sent: Wednesday, May 14, 2008 6:43 PM Subject: [stunnel-users] Reg:IPV6 support on stunnel Hi, I am newbie to stunnel.As mentioned in some of your posts i just gone through ..how to enable ipv6 support. While configuring i did the same by giving the following command. ./configure --enable-ipv6 make make install and i ran stunnel which listens at some port and i have my own server which listens for stunnel. I have enabled IPV6 ip When my client tries to connect to my ipv6 ip and the port(stunnel listens) where my it is listening...Its failed to connect. Could you please throw some light on how to enable IPV6 support for stunnel apart from what i tried. Thanks in advance Sandhya -------------------------------------------------------------------------------- _______________________________________________ stunnel-users mailing list stunnel-users(a)mirt.net http://stunnel.mirt.net/mailman/listinfo/stunnel-users

1 0

Reg:IPV6 support on stunnel
by Sandhya 14 May '08

14 May '08

Hi, I am newbie to stunnel.As mentioned in some of your posts i just gone through ..how to enable ipv6 support. While configuring i did the same by giving the following command. ./configure --enable-ipv6 make make install and i ran stunnel which listens at some port and i have my own server which listens for stunnel. I have enabled IPV6 ip When my client tries to connect to my ipv6 ip and the port(stunnel listens) where my it is listening...Its failed to connect. Could you please throw some light on how to enable IPV6 support for stunnel apart from what i tried. Thanks in advance Sandhya

1 0

Reg:IPV6 support on stunnel
by Sandhya 13 May '08

13 May '08

Hi, I am newbie to stunnel.As mentioned in some of your posts i just gone through ..how to enable ipv6 support. While configuring i did the same by giving the following command. ./configure --enable-ipv6 make make install and i ran stunnel which listens at some port and i have my own server which listens for stunnel. I have enabled IPV6 ip When my client tries to connect to my ipv6 ip and the port(stunnel listens) where my it is listening...Its failed to connect. Could you please throw some light on how to enable IPV6 support for stunnel apart from what i tried. Thanks in advance Sandhya

1 0

Re: [stunnel-users] HTTPS connections aborted
by Matthias Kellermann 09 May '08

09 May '08

Hi Aleksandar, Aleksandar Lazic wrote: > The timing events at 6st position. > > http://haproxy.1wt.eu/download/1.3/doc/haproxy-en.txt > > > --- > 4.2.3) Timing events > -------------------- > . > . > . > > Other cases ('xx' means any value to be ignored) : > -1/xx/xx/xx/Tt: the client was not able to send its complete request > in time, > or that it aborted it too early. > Tq/-1/xx/xx/Tt: it was not possible to process the request, maybe because > servers were out of order. > Tq/Tw/-1/xx/Tt: the connection could not establish on the server. > Either it > refused it or it timed out after Tt-(Tq+Tw) ms. > Tq/Tw/Tc/-1/Tt: the server has accepted the connection but did not > return a > complete response in time, or it closed its connexion > unexpectedly, after Tt-(Tq+Tw+Tc) ms. thanks, I'll have a look at it. -Matthias

1 0

Feature request - verify fall-back
by Sudhaker Raj 09 May '08

09 May '08

Hi Team, I wish to use stunnel for following use-case (to create a highly-protected website which can be accessed only using a valid client-cert). gateway.example.com:443 -> public.example.com:80 (when client-cert verification fails) gateway.example.com:443 -> intranet.example.com:80 (when client-cert verification ok - normally hidden from public) As of now stunnel simply drops the connection when service is configured to verify the client certificate and verification fails. Is it possible to add a fall-back connect when verification fails. [protected-web] verify=3 accept=443 connect=intranet.example.com:80 noverify=public.example.com:80 I guess it will be a nice addition to stunnel's feature list. Thanks, Sudhaker

3 3

launch stunnel server daemon from php?
by Wesley Kenzie 05 May '08

05 May '08

I am having problems launching stunnel from a php "exec" command. I would like to have it launch as a server daemon and then have the exec command return control to the php script. The best I have been able to do is launch it in foreground with a command like "/usr/local/bin/stunnel /etc/stunnel/test.conf &" but control never returns to the php script. I do not want to use inetd mode, and would prefer to specify command line arguments, but I see they are no longer available in version 4. Any ideas? . . . . . Wesley Kenzie wkenzie(a)gmail.com

2 2

HTTPS connections aborted
by Matthias Kellermann 05 May '08

05 May '08

Hi list, I've set up a loadbalancer with haproxy and I'm using stunnel to have HTTPS on the loadbalancer. Every connection on port 443 is forwarded by stunnel to port 80 where haproxy listens. Some facts: FreeBSD 6.2 i386 stunnel 4.22 OpenSSL 0.9.7e-p1 gcc version 3.4.6 [FreeBSD] 20060305 Here is my stunnel.conf: sslVersion = all chroot = /var/tmp/stunnel setuid = stunnel setgid = nogroup pid = /stunnel.pid socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 debug = 3 [https] cert = /usr/local/etc/stunnel/ssl/mycert.pem accept=localip:443 connect=localip:80 Some of my customers using HTTPS are claiming about aborted connections with MS Internet Explorer. They get a message about "secure and unsecure objects on this site, proceed?". When clicking yes they get a IE error message: "site not found". (Sorry, I don't know the exact english error message because customers use german version of IE). Of course there are only HTTPS objects on the page. In my logs I've lot of these entries: SSL_accept: Peer suddenly disconnected From time to time I'll get this error: SSL_read: Connection reset by peer (54) or: SSL_read: Operation timed out (60) or: SSL_write: Broken pipe (32) At the moment I have no idea where these errors come from and if they really belong to the error the customer gets with his IE browser. Any ideas what could cause these error messages? Thanks in advance, Matthias

2 1

Stunnel <=> p2p
by mluft 05 May '08

05 May '08

Hi, i just started working with stunnel, and i'm really impressed! But i've got 2 questions: * How is it realized, that a setup like this: accept = 15000 connect = 16000 also encrypts the reply of the app which listens on port 16000? Is there a kind of masquerading? * I think of a tool like tsocks or socksify: "tsocks application" redirects all traffic from application over an socks proxy. Is it possible to realize such a redirection with stunnel? So that all outgoing traffic of an application gets encrypted? I know that i can build point-to-point tunnels, but that's not possible for my scenario... Thanks & Cheers, Matthias

1 0

Stunnel 4.23 xforwardfor patch
by Rene Plattner 05 May '08

05 May '08

Dear stunnel users! I noticed that the xforwardfor patch does not work any more on version > 4.20! This is due an function renaming in stunnel! After adding xforwardfor patch for version 4.20 see options.c: Simply replace log_raw(...); to s_log(LOG_RAW,...); chears, -- Dipl.-Ing. Rene' Plattner Zentraler Informatikdienst (ZID) Universität Innsbruck, Österreich Technikerstr. 13 A-6020 Innsbruck Tel: ++43512/507-2360 Fax: ++43512/507-2944

1 0

Can stunnel be configured to listen for multiple IP addresses and access different email hosts?
by F Boyd Crow 03 May '08

03 May '08

I have attempted to use stunnel to retrieve POP3 SSL mail from two different ISPs: pop.att.yahoo.com and pop.gmail.com. I have tried putting in separate Access/Connect line pairs under [pop3s] either point to the different ports on 127.0.0.1 whch is an Exchange Server 2003 machine. I have tried pointing both to the same port. In either case, the first POP3 host is accessed but not the second. I have also tried one Access line and two separate connects. It works fine with one Access line and one Connect line but a previous post on this subject indicated I should be able to do retrievals from two separate sites and I do not see how. I am using the POP3 Connection Manager on Small Business Server 2003. I set the email server to 127.0.0.1 and use port 9995. The corresponding section of the stunnel.conf file is client = yes debug = 7 ; Service-level configuration [pop3s] accept = 9995 connect = pop.gmail.com:995 ;does not work with two separate pop3s, either with two separate accept ports or one common final comment is my own. Has anyone successfully accessed mail from two POP3 servers through stunnel or have any suggestions? Boyd

1 0

stunnel 4.23 released
by Michal Trojnara 03 May '08

03 May '08

Dear Users, I have just released a new version of stunnel. Please find below the ChangeLog entry: Version 4.23, 2008.05.03, urgency: HIGH: * Bugfixes - Local privilege escalation bug on Windows NT based systems fixed. A local user could exploit stunnel running as a service to gain localsystem privileges. Home page/download: http://stunnel.mirt.net/ sha1sum for stunnel-4.23.tar.gz file: d0fef8b518a44b9623692381a53680e0b4b01686 Best regards, Mike

1 0

Re: [stunnel-users] launch stunnel server daemon from php?
by Wesley Kenzie 02 May '08

02 May '08

No, that does not work either. I can get stunnel to start fine outside of php, but not as an exec or shell_exec from within a running script, or even from within php -a interactively. I can get other programs to work fine this way, so it is specific to the stunnel executable as best I can tell. I have tried running various forms of bash scripts too, but none do any better. Ideas? . . . . . Wesley On Fri, May 2, 2008 at 12:52 PM, Ludolf Holzheid < lholzheid(a)bihl-wiedemann.de> wrote: > On Fri, 2008-05-02 12:32:19 -0700, Wesley Kenzie wrote: > > Yes. Without this stunnel would not launch. With foreground = no and > with > > the trailing & on the command line I got it to launch, but still control > > will not return to my php script ... > > > > . . . . . > > Wesley > > > > On Fri, May 2, 2008 at 12:29 PM, Ludolf Holzheid < > > lholzheid(a)bihl-wiedemann.de> wrote: > > > > > On Fri, 2008-05-02 12:19:38 -0700, Wesley Kenzie wrote: > > > > I am having problems launching stunnel from a php "exec" command. I > > > would > > > > like to have it launch as a server daemon and then have the exec > command > > > > return control to the php script. > > > > > > Did you try to specify ``foreground = no'' in the configuration file? > > Wesley, > > I must confess that I don't know php and its `exec' command. There > must be something special about it. > > I would try to start a shell, which starts stunnel, which, in turn, > detaches itself to the background (i.e. something like "/bin/sh -c > '/usr/local/bin/stunnel /etc/stunnel/test.conf'"). So that the process > started by `exec' finishes and hopefully control returns to php. > > hth, > > Ludolf > > -- > > A: Because it messes up the order in which people normally read text. > Q: Why is top-posting such a bad thing? > A: Top-posting. > Q: What is the most annoying thing in e-mail? >

1 0

Re: [stunnel-users] launch stunnel server daemon from php?
by Wesley Kenzie 02 May '08

02 May '08

Yes. Without this stunnel would not launch. With foreground = no and with the trailing & on the command line I got it to launch, but still control will not return to my php script ... . . . . . Wesley On Fri, May 2, 2008 at 12:29 PM, Ludolf Holzheid < lholzheid(a)bihl-wiedemann.de> wrote: > On Fri, 2008-05-02 12:19:38 -0700, Wesley Kenzie wrote: > > I am having problems launching stunnel from a php "exec" command. I > would > > like to have it launch as a server daemon and then have the exec command > > return control to the php script. > > Did you try to specify ``foreground = no'' in the configuration file? > > Ludolf > > -- > > --------------------------------------------------------------- > Ludolf Holzheid Tel: +49 621 339960 > Bihl+Wiedemann GmbH Fax: +49 621 3392239 > Floßwörthstraße 41 e-mail: lholzheid(a)bihl-wiedemann.de > D-68199 Mannheim, Germany > --------------------------------------------------------------- >

1 0

Verify=3 restart needed ?
by Edouard Dessioux 30 Apr '08

30 Apr '08

Hi everyone, Sorry for the repost, but in the web mailing list, my previous message appears weird, so I thought I could re-post it. I'm using stunnel v4.16 on a Windows 2003 Server, and I'm working with stunnel in verify=3 mode. I wanted to know if the stunnel needs to be restarted after a certificates has been removed ? If not, how long is the certificate considered valid after the file has been removed ? Regards, Edouard DESSIOUX Directeur de Projets Tibco Mobile 3, rue Danton - 92240 Malakoff Tél : +33 (0)1 55 58 04 59 - Fax : +33 (0)1 55 58 03 89 - Mob. +33 (0)6 34 02 61 54 E-mail : edessioux(a)tibco.fr - www.tibcomobile.fr Faites un geste pour la planète, n'imprimez ce message que si nécessaire.

3 3

to use certificate and key file in another machine
by Xiaoming Dong 29 Apr '08

29 Apr '08

Stunnel Version 4.2.1 Run in windows 2k3 When I configured stunnel to use cert from another windows machine (I already mapped the drive to this machine). Restarting stunnel got following error: 2008.04.29 10:01:32 LOG7[3568:1792]: Snagged 64 random bytes from C:/.rnd 2008.04.29 10:01:32 LOG7[3568:1792]: Wrote 1024 new random bytes to C:/.rnd 2008.04.29 10:01:32 LOG7[3568:1792]: RAND_status claims sufficient entropy for the PRNG 2008.04.29 10:01:32 LOG7[3568:1792]: PRNG seeded successfully 2008.04.29 10:01:32 LOG3[3568:1792]: V:\Program Files\client_cert\private\client_key.pem: No such process (3) 2008.04.29 10:01:32 LOG3[3568:1792]: Server is down Does anybody have any clue? Thanks, -Xiaoming ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

1 0

Will fwknop work?
by richard.woodman＠cox.net 29 Apr '08

29 Apr '08

If I understand the question correctly, isn't this what "port knocking" or single packet authorization (e.g. fwknop) is supposed to do? I have used fwknop and SSH in our lab, but only with Linux and iptables. However, I think fwknop is supposed to interface with more than just iptables on the local box (meaning you would not have to use a Linux box to replace your current firewall). I think you can use fwknop to monitor syslog and parse for specific events and then open the port. In other words, your current firewall reports to your syslog server and fwknop parses the log file for the security event associated with the reception of a SPA packet on your outside interface. Fwknop then sends your firewall (through a script?) whatever command is required to open the port you want and redirect it to the appropriate inside machine (or you could simply enable / disable a preconfigured rule). I am not a scripting guru so I may be WAY off base here and if I am, I apologize for leading you astray. Anyway, you might want to check out the following: http://cipherdyne.org/fwknop/ --> FireWall KNock Operator home page http://fwknop.darwinports.com/ --> OS X fwknop client There is also a Windows UI version that is supposed to create SPA packets without using fwknop / PERL or running under Cygwin but I have not used that. Richard On 4/29/08 7:50 PM, "jz(a)ellingtongeologic.com" <jz(a)ellingtongeologic.com> wrote: > > Good Morning Mike: > > I had a question and sent to the list (it might have not gone thru) The > question was that: is it possible for stunnel to go to the router, for > example, 10.10.1.1, to scan for a port of interest and see whether there is a > request thru that port? so the nat router would not have to forward the port > to the stunnel of my local machine, e.g. 10.10.1.188, on which stunnel is > listening for port 8888 and will relay it to 5631 of the local program. > > Thanks

1 0

how to enable TLS?
by Zhang Weiwu 26 Apr '08

26 Apr '08

Dear all I just configured stunnel for pop+ssl which works fine. What I want is actually smtp+tls. Is there a way to do so? I tried to configure thunderbird to access the port stunnel is listening to, with TLS. Then I got "Connectioin to xxx.xxx.xxx timed out". Best regards -- Real Softservice Huateng Tower, Unit 1788 Jia 302 3rd area of Jinsong, Chao Yang Tel: +86 (10) 8773 0650 ext 603 Mobile: 135 9950 2413 http://www.realss.com

1 0

Query regarding stunnel performance
by Arati.Kumar＠ubs.com 25 Apr '08

25 Apr '08

Hi, Does anyone know if there are any known issue with stunnel 4.05 and oracle 10g (especially in retrieving blobs) ? . Our application uses stunnel to communicate with oracle database. We are currently experiencing massive performance degradation after upgrading oracle from 9i to 10g (almost 6 times slower from what it was in 9i). Our application stores some data in blobs this seems to be the root cause of slower performance. Pls can someone let me know if there are any known issues with stunnel and 10g blob retrieving ? Also If an upgrade is recommended, which is the stable version we must be switching to? Thanks, Ararti Visit our website at http://www.ubs.com This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. E-mails are not encrypted and cannot be guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or contain viruses. The sender therefore does not accept liability for any errors or omissions in the contents of this message which arise as a result of e-mail transmission. If verification is required please request a hard-copy version. This message is provided for informational purposes and should not be construed as a solicitation or offer to buy or sell any securities or related financial instruments. UBS Limited is a company registered in England & Wales under company number 2035362, whose registered office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom. UBS AG (London Branch) is registered as a branch of a foreign company under number BR004507, whose registered office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom. UBS Clearing and Execution Services Limited is a company registered in England & Wales under company number 03123037, whose registered office is at 1 Finsbury Avenue, London, EC2M 2PP, United Kingdom.

1 0

Linux FIPS compile libary question
by Joe Kemp 23 Apr '08

23 Apr '08

I am compiling stunnel on Centos 5 that has a regular Openssl 0.9.8b rpm installed. I have put my FIPS openssl in /usr/local/sslfips112. Configure with: ./configure --with-ssl=/usr/local/sslfips112 --enable-fips --disable-libwrap Make's linker line: /bin/sh ../libtool --tag=CC --mode=link FIPSLD_CC=gcc /usr/local/sslfips112/bin/fipsld -g -O2 -Wall -Wshadow -Wcast-align -Wpointer-arith -I/usr/local/sslfips112/include -lldap -o stunnel file.o client.o log.o options.o protocol.o network.o resolver.o ssl.o ctx.o verify.o sthreads.o stunnel.o auth.o pty.o libwrap.o -lz -ldl -lutil -lnsl -lpthread -L/usr/local/sslfips112/lib -lssl -lcrypto FIPSLD_CC=gcc /usr/local/sslfips112/bin/fipsld -g -O2 -Wall -Wshadow -Wcast-align -Wpointer-arith -I/usr/local/sslfips112/include -o stunnel file.o client.o log.o options.o protocol.o network.o resolver.o ssl.o ctx.o verify.o sthreads.o stunnel.o auth.o pty.o libwrap.o -lldap -lz -ldl -lutil -lnsl -lpthread -L/usr/local/sslfips112/lib -lssl -lcrypto This builds a stunnel that seems to run fine. During startup it says "stunnel is in FIPS mode." But if I run "ldd stunnel" it shows it needs /lib/libssl.so.6. While stunnel is running lsof shows it has that library open also. Why does my FIPS stunnel build still use the 0.9.8b shared library? Shouldn't all of the ssl dependencies been handled by the static FIPS openssl library during linking? The same issue exists for libcrypt.

4 6

stunnel wildcard question
by neve_capricorn＠gmx.de 17 Apr '08

17 Apr '08

hi, iam pulling my hair out with an wildcard ssl certificate / stunnel configuration. i use stunnel 4.2.2 and latest openSSL release. i have a signed wildcard certificate from comodo. when i add the .crt, .key and .ca-bundle file into an apache configuration, everything work fine in firefox and ie. when i try to use stunnel with the same certificates i got problems in firefox when i disable ; Authentication stuff ; verify =1 and when i set verify=1 i got problems in IE .7 (a dialog box pops and ask for identification the certificates). can someone maybe help me out,please? a demo is here: https://ssl1.medialib.de (works fine in firefox) https://ssl1.medialib.de (problem in IE7) thank you very much & best gary

1 0

Verify=3 cache expiration
by Edouard Dessioux 16 Apr '08

16 Apr '08

Hi everyone, I'm using stunnel v4.16 on a Windows 2003 Server, and I'm working with stunnel in verify=3 mode. I wanted to know if the stunnel needs to be restarted after a certificates has been removed ? If not, how long is the cache expiration delay before being taken into account ? Regards, Edouard DESSIOUX Directeur de Projets Tibco Mobile 3, rue Danton - 92240 Malakoff Tél : +33 (0)1 55 58 04 59 - Fax : +33 (0)1 55 58 03 89 - Mob. +33 (0)6 34 02 61 54 E-mail : edessioux(a)tibco.fr - www.tibcomobile.fr Faites un geste pour la planète, n'imprimez ce message que si nécessaire.

1 0

SCM repo?
by Alberto Giménez 12 Apr '08

12 Apr '08

Hello, Is there any public SCM (SVN, GIT, whatever) repository? I haven't seen anything on the website and I'd like to try to implement a new feature. Thanks in advance. -- Alberto Giménez

2 1

Stunnel 4.22 compiling issues on Solaris 10
by cod3fr3ak 09 Apr '08

09 Apr '08

I have sucessfully compiled Stunnel 4.22 on Solaris 10, but still recieve the sam errors during the configure. Initially I tried using Openssl-0.9.8f from Sun Freeware. Then I built and compiled Openssl-0.9.8g, but I still recieved the following errors on the build: configure: **************************************** SSL checking for SSL directory... /usr/local/ssl checking for obsolete RSAref library... no checking /usr/local/ssl/include/openssl/engine.h usability... yes checking /usr/local/ssl/include/openssl/engine.h presence... no configure: WARNING: /usr/local/ssl/include/openssl/engine.h: accepted by the compiler, rejected by the preprocessor! configure: WARNING: /usr/local/ssl/include/openssl/engine.h: proceeding with the compiler's result checking for /usr/local/ssl/include/openssl/engine.h... yes <ftp://ftp.sunfreeware.com/pub/freeware/sparc/10/openssl-0.9.8f-sol10-sparc-…>

1 0

Using stunnel + haproxy for SSL support
by Alberto Giménez 04 Apr '08

04 Apr '08

Hi, I'm having trouble using stunnel and haproxy to load balance https and http traffic. To be honest, I really don't know wether it is stunnel or haproxy related so I am going to contact both lists :) I have set up a haproxy load balancer as http proxy for two backend Apache2 webservers. It works fine. I also have stunnel on the same LB to add SSL suport (with xforwardedfor patch installed). It works fine (mostly). The issue is related to Apache trailing slash thingie. If I query https://haproxy.domain.loc/hatest/ it works perfectly, but if I omit the trailing slash: https://haproxy.domain.loc/hatest then following things happen: - Browser makes SSL connection with stunnel on port 443. - stunnel deciphers and forwards the request on the haproxy attached to LB's port 80. - haproxy (now using plain http) forwards to one of the backends. - Apache2 located on the backend replies with "301 moved permanently" to force the browser to add the trailing slash. As Apache was queried by *haproxy in plain http*, the 301 includes http:// on the Location header. HTTPS is over from now! - The client browser then rewrites the address to http://haproxy.domain.loc/hatest/ and SSL is lost forever. I've been googling and searching the lists but nothing found, just this old message: http://mirt.net/pipermail/stunnel-users/2007-January/001437.html Has anyone found a workaround for that issue? Thanks in advance. -- Alberto Giménez

2 1

Problems with Stunnel 4.22 build
by GGrant 04 Apr '08

04 Apr '08

There's a problem with the Stunnel 4.22 build as there's undefined IDs in the resources.rc file. To get around this get the resources.rc file from an earlier build (e.g. Stunnel 4.21) and replace it. It should compile happily -- View this message in context: http://www.nabble.com/Problems-with-Stunnel-4.22-build-tp16491431p16491431.… Sent from the Stunnel - Users mailing list archive at Nabble.com.

1 0

23 secs to list 5000 files using stunnel and samba on linux
by Paul Kerin 04 Apr '08

04 Apr '08

Hi I'm using the following s/w versions on RHEL3: Stunnel: 4.04 (also experimenting with 4.22, no difference so far) Samba: 3.0.9-1.3E.10 Kernel: 2.4.21-32.0.1.EL Stunnel is used to encrypt samba connections from linux clients because redhat 3 doesn't support NTLMv2 (cifs not standard and probably not an option). To list 5000 files (simply typing ls within a mounted directory on the client) it consistently takes around 20-23 seconds to return the data. Listing is almost instantaneous when using a windows client, or using a linux client without stunnel. As a side note, if I pipe the result to /dev/null it takes around 9 seconds (?). The network forwarding path from the samba client to samba server is: smbclient > localhost:924 > stunnel > remotehost:923 > stunnel > samba server (port 446) Tcpdump shows that when using stunnel about 10500 packets are generated, minus stunnel it's more like 500. I'd expect some overhead related to SSL, but 21 times the traffic seems a little excessive. I've experimented with socket options such as TCP_NODELAY, SO_LINGER, SO_RCVLOWAT, SO_OOBINLINE, etc with no improvement at all. However, my understanding of these is pretty superficial so I'm not confident I've exhausted all options here (ie perhaps combining multiple settings at once). I've got stunnel debug set to 7 on client and server. No errors and no logging at all except for the initial handshake when the mount is created. Including the tcpdump would probably be excessive at this stage. In summary, using stunnel the data gets transmitted in packets usually containing around 200 bytes, whereas without stunnel it's mostly 1408 byte packets. Any suggestions? Thanks Paul

4 4

Stunnel Newbie - Compiling Stunnel and adding patches for Windows
by GGrant 02 Apr '08

02 Apr '08

Hi, I've downloaded the Stunnel source code and I am trying to compile it with the Visual C++ compiler (cl.exe). I am not sure what to do as I cannot find the Makefile.w32. So how do I go about it? Also how do you apply a user patch to the source code/ Must I copy and paste the code to the required files? Cheers -- View this message in context: http://www.nabble.com/Stunnel-Newbie---Compiling-Stunnel-and-adding-patches… Sent from the Stunnel - Users mailing list archive at Nabble.com.

1 0

Stunnel-Openssl: Problems with certificate chain
by Pietro Di Primo 31 Mar '08

31 Mar '08

Hi, My name is Pietro Di Primo and I work on the INFN Grid project (http://grid.infn.it/). I'm trying to develop an application based on stunnel. Everything works well when I use user certificates (signed by our CA), but i need to use (on the client side) a particular proxy. This proxy contains a public and a private key, and the public key of the user that generated it (I atteched a file with a such proxy). On the server side I have the CA certificate but not the user's one, so I can't verify the client (unable to get local issuer certificate). I tried to get the user certificate using the SSL_get_peer_cert_chain function on the server, but it returns NULL. On the client side I tried SSL_CTX_use_certificate_chain_file, Also I tried to get a STACK_OF(X509) and use SSL_CTX_use_certificate (for proxy certificate), and SSL_CTX_add_extra_chain_cert (to include the user's certificate), but it still doesn't work. Can you help me please? Best Regards -- --- Ing. Pietro Di Primo INFN - Istituto Nazionale di Fisica Nucleare Grid Computing Research Area Via S. Sofia 78 95123 CATANIA (Italy) Voice : +39 095 378-5446/5517 Fax. : +39 095 378 5518 Mobile : +39 320 8591527 E-mail : pietro.diprimo(a)ct.infn.it Skype : pietrodiprimo

1 0

Re: [stunnel-users] Enhancement request - Increase thread stack size
by Mccue, Richard Alan 26 Mar '08

26 Mar '08

Michal Trojnara wrote: >> I load an engine into Stunnel to access the private key. The engine loads >> without incident, but Stunnel consistently aborts when creating a session >> with the first peer because the thread stack is too small. > > I'm not sure if it's a good idea to sacrifice stunnel scalability for > compatibility with this engine. Thread stack size is one of the main > limitations for the number of concurrent threads on 32-bit platforms. IMHO > it's better to fix the engine code to allocate its buffers on the heap.> > > What do you think? All things considered, I must agree that reducing the engine use of the stack would be the better solution. I'll look it over and see if I can find anything sizable on the stack to move. Thanks for the suggestion! - Rich

2 2

New libwrap code
by Randy McMurchy 26 Mar '08

26 Mar '08

Hi all, My apologies if this has been discussed (I've reviewed the archives since 4.21 was released), and if so please just point me in the right direction. I read in some of the messages there's been a revamp (overhaul, changes, whatever) of the libwrap code in Stunnel-4.21. I'm not having any issues using the new version, however, I'm seeing it spawn 5 processes where in every other version I've used in the past there has always only been one process spawned. Here's what I see: root 3408 1 0 17:50 pts/10 00:00:00 /usr/bin/stunnel root 3409 1 0 17:50 pts/10 00:00:00 /usr/bin/stunnel root 3410 1 0 17:50 pts/10 00:00:00 /usr/bin/stunnel root 3411 1 0 17:50 pts/10 00:00:00 /usr/bin/stunnel root 3412 1 0 17:50 pts/10 00:00:00 /usr/bin/stunnel stunnel 3413 1 0 17:50 ? 00:00:00 /usr/bin/stunnel Is this a product of the new libwrap code? I turned debug messages WAY up and I see over and over startup messages for all 5 processes. Example of one: 2008.03.25 17:50:24 LOG5[3407:3082803904]: stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8e 23 Feb 2007 2008.03.25 17:50:24 LOG5[3407:3082803904]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv4 Auth:LIBWRAP 2008.03.25 17:50:24 LOG6[3407:3082803904]: file ulimit = 1024 (can be changed with 'ulimit -n') 2008.03.25 17:50:24 LOG6[3407:3082803904]: poll() used - no FD_SETSIZE limit for file descriptors 2008.03.25 17:50:24 LOG5[3407:3082803904]: 500 clients allowed 2008.03.25 17:50:24 LOG7[3407:3082803904]: FD 9 in non-blocking mode 2008.03.25 17:50:24 LOG7[3407:3082803904]: FD 10 in non-blocking mode 2008.03.25 17:50:24 LOG7[3407:3082803904]: FD 11 in non-blocking mode 2008.03.25 17:50:24 LOG7[3407:3082803904]: SO_REUSEADDR option set on accept socket 2008.03.25 17:50:24 LOG7[3407:3082803904]: swat bound to 0.0.0.0:904 2008.03.25 17:50:24 LOG7[3413:3082803904]: Created pid file /run/stunnel.pid 2008.03.25 17:50:42 LOG7[3413:3082803904]: swat accepted FD=12 from 192.168.11.212:1535 2008.03.25 17:50:42 LOG7[3413:3086076816]: swat started 2008.03.25 17:50:42 LOG7[3413:3086076816]: FD 12 in non-blocking mode 2008.03.25 17:50:42 LOG7[3413:3086076816]: Waiting for a libwrap process 2008.03.25 17:50:42 LOG7[3413:3086076816]: Acquired libwrap process #0 2008.03.25 17:50:42 LOG7[3413:3086076816]: Releasing libwrap process #0 2008.03.25 17:50:42 LOG7[3413:3086076816]: Released libwrap process #0 2008.03.25 17:50:42 LOG7[3413:3086076816]: swat permitted by libwrap from 192.168.11.212:1535 2008.03.25 17:50:42 LOG5[3413:3086076816]: swat accepted connection from 192.168.11.212:1535 When I connect to swat (Samba config tool) it again goes through a process of using all 5 of the system processes. Sorry for being so verbose here. My question is: Has something changed so that there is more than one process spawned (and used) when Stunnel is started? Any help would be appreciated. -- Randy

1 0

Enhancement request - Increase thread stack size
by Mccue, Richard Alan 25 Mar '08

25 Mar '08

I load an engine into Stunnel to access the private key. The engine loads without incident, but Stunnel consistently aborts when creating a session with the first peer because the thread stack is too small. I rebuilt Stunnel 4.21 on my test machine after increasing the stack size parameter and it fixed the problem. Here is the diff: --------------------- # diff -b common.h common.h_orig 33c33 < #define STACK_SIZE 32768*4 --- > #define STACK_SIZE 32768 42c42 < #define STACK_SIZE 65536*4 --- > #define STACK_SIZE 65536 --------------------- Could this enhancement be included in the next Stunnel release? Regards, Rich

1 0

HTTPS being redirected back to HTTP
by Steven Buehler 25 Mar '08

25 Mar '08

I have tested several versions of stunnel up to version 4.21 on SLES 9 and 10. I am trying to use stunnel to implement TLS 1.0 for HTTPS on a Lotus Domino server. I am able to connect to the Domino server with a web browser with HTTPS successfully. However, once I log into the server it drops me back down to HTTP. I suspect the Domino server is sending a redirect back to HTTP during the login because if I change the URL back to HTTPS in the browser after the login, it re-establishes the HTTPS connection. I've checked with Lotus support and there is nothing that can be done on the Domino server to prevent this. Is there anyway to handle this in stunnel so that the requests remain as HTTPS requests to the browser? Here is the services section of my stunnel.conf [https] sslVersion = TLSv1 accept = gsvatestm02.er.usgs.gov:443 connect = gsvatestm02.er.usgs.gov:80 TIMEOUTclose = 0

2 1

possible stunnel memory leak
by Stefano Stabellini 25 Mar '08

25 Mar '08

Hi all, I may have found a possible memory leak in stunnel (version 4.15) when used with pthreads. What I am seeing is the memory usage going up when connections open and never going down, no matter if all the connections close. It is really easy to reproduce it: I have started two stunnel instances communicating with each other, the first one acting as a server (config file stunnel.1) and executing cat, the second one acting as a client (config file stunnel.2) and opening port 5556. Then I execute the script I have attached that simply opens 500 connections using netcat to port 5556. If you cat /proc/<stunnel pid>/status you can see that the memory usage is about 300MB. The problem is that even after all the connections are closed the memory usage stays (more or less) the same. I tried to find the bug myself using valgrind, but I couldn't get far, it seems that valgrind gets confused by openssl so doesn't print very useful informations. I really appreciate any help you can give me; just tell me if you need any other info. Best Regards, Stefano Stabellini #!/bin/bash i=1 while test $i -lt 520 do cat script | nc 127.0.0.1 5556 &>/dev/null & echo $i i=$(( $i + 1 )) done foreground = yes [test] accept = 5555 cert = /etc/ssl/certs/stunnel.pem exec = cat foreground = yes [test] accept = 5556 cert = /etc/ssl/certs/stunnel.pem client = yes connect = 5555

1 0

IMAPS stunnel on older OSX Mac: SSL3_GET_RECORD:wrong version number
by Tom Shaw 23 Mar '08

23 Mar '08

1. compiled and installed stunnel on older OSX Mac. Tried connectiong via IMAPS and POP3S using Apple OSX Mail and I get the errors shown in line 4. Process works when I change my .conf file to contain sslVersion = SSLv2. I'm a neophyte to ssl and I have no idea what is wrong or where to start or if this situation is OK. Any help is appreciated. Also this needs to be backward compatible to OSX 10.3.9 which is why the compiling on Panther. 2. stunnel version 4.20 3. standalone via sudo from command line 4. error message Mar 23 10:49:52 musky stunnel: LOG5[2941:25192960]: imaps accepted connection from 206.208.58.127:37295 Mar 23 10:49:52 musky stunnel: LOG3[2941:25192960]: SSL_accept: 1408F10B: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number Mar 23 10:49:52 musky stunnel: LOG5[2941:25192960]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 5. $ /usr/local/sbin/stunnel -version stunnel 4.20 on powerpc-apple-darwin7.9.0 with OpenSSL 0.9.7l 28 Sep 2006 Threading:PTHREAD SSL:ENGINE Sockets:SELECT,IPv4 Auth:LIBWRAP Global options debug = 5 pid = /usr/local/var/run/stunnel/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /usr/local/etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /usr/local/etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none 6. $ uname -a Darwin musky.oitc.com 7.9.0 Darwin Kernel Version 7.9.0: Wed Mar 30 20:11:17 PST 2005; root:xnu/xnu-517.12.7.obj~1/RELEASE_PPC Power Macintosh powerpc 7. Your libc version if you use Linux. 8. $ gcc -v Reading specs from /usr/libexec/gcc/darwin/ppc/3.3/specs Thread model: posix gcc version 3.3 20030304 (Apple Computer, Inc. build 1666) 9. $ openssl version OpenSSL 0.9.7l 28 Sep 2006 10. My conf ; Sample stunnel configuration file by Michal Trojnara 2002-2006 ; Some options used here may not be adequate for your particular configuration ; Please make sure you understand them (especially the effect of chroot jail) ; Certificate/key is needed in server mode and optional in client mode cert = /usr/local/etc/stunnel/secure_mail.pem ;key = /usr/local/etc/stunnel/secure_mail.pem ; Protocol version (all, SSLv2, SSLv3, TLSv1) sslVersion = SSLv2 ; Some security enhancements for UNIX systems - comment them out on Win32 ;chroot = /usr/local/var/lib/stunnel/ ;setuid = nobody ;setgid = nobody ;setgid = nogroup ; PID is created inside chroot jail pid = /var/run/stunnel.pid ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ;compression = rle ; Workaround for Eudora bug options = DONT_INSERT_EMPTY_FRAGMENTS ; Authentication stuff ;verify = 2 ; Don't forget to c_rehash CApath ; CApath is located inside chroot jail ;CApath = /certs ; It's often easier to use CAfile ;CAfile = /usr/local/etc/stunnel/certs.pem ; Don't forget to c_rehash CRLpath ; CRLpath is located inside chroot jail ;CRLpath = /crls ; Alternatively you can use CRLfile ;CRLfile = /usr/local/etc/stunnel/crls.pem ; Some debugging stuff useful for troubleshooting debug = 7 ;output = stunnel.log ; Use it for client mode ;client = yes ; Service-level configuration [pop3s] accept = 995 connect = 110 [imaps] accept = 993 connect = 143 [ssmtp] accept = 465 connect = 587 [sldap] accept = 636 connect = 389 ;[https] ;accept = 443 ;connect = 80 ;TIMEOUTclose = 0 ; vim:ft=dosini

1 0

What is the stable version of Stunnel?
by Randy McMurchy 22 Mar '08

22 Mar '08

Hi all, Can anyone say for sure what is the stable version of Stunnel? On the main Stunnel web site page ( http://www.stunnel.org ) and in the announcement for the version it says the new 4.21 version is experimental. However, on the download page ( http://www.stunnel.org/download/source.html ) the 4.21 version is flagged as the stable version. Sort of confusing. -- Randy

2 1

compiling stunnel 4.21 on solaris 10
by cod3fr3ak 14 Mar '08

14 Mar '08

Is this not compatible with Openssl 0.9.8f? I think 0.9.8g only reacently came out. Is this the only thing wrong. All my other applications work with this version of openssl. BTW I am using the sunfreeware version, because I wanted to avoid having to compile a bunch of software. i was planning to only compile Stunnel, and maybe swatch. I'll compile Openssl if there is no other way but I'd rather not. Any ideas? R

1 0

bidirectional tunnel
by Leo Cherepanov 14 Mar '08

14 Mar '08

Hello. First of all - thanx a lot, I'm using stunnel every day and it helps me. It will help me even more :) if I can make bidirectional tunnel between external server S and computer C behind nat-proxy-firewall. Of course, C is initiator. I mean after tunnel is up, there is a port-forward C:port1->S:port2 (already works) _and_ S:port3->C:port4 Once again, S has an external ip, C has a private ip. Is this schema posible? Thanx en avance.

1 0

SSLV3 decompression failure???
by Earl Stenlund 13 Mar '08

13 Mar '08

I've been running into the error below: SSL_connect: 14094406: error:14094406:SSL routines:SSL3_READ_BYTES:sslv3 alert decompression failure Running Vista SP1 with stunnel 4.21 connecting to newshosting. It seems to come and go. Any ideas on what is causing this or how to fix it? Thanks, Earl

1 0

Re: [stunnel-users] Compiling Stunnel 4.21 on Solaris 10
by cod3fr3ak 13 Mar '08

13 Mar '08

Is this not compatible with Openssl 0.9.8f? I think 0.9.8g only reacently came out. Is this the only thing wrong. All my other applications work with this version of openssl. BTW I am using the sunfreeware version, because I wanted to avoid having to compile a bunch of software. i was planning to only compile Stunnel, and maybe swatch. I'll compile Openssl if there is no other way but I'd rather not. Any ideas? R

1 0

Compiling Stunnel 4.21 on Solaris 10
by cod3fr3ak 13 Mar '08

13 Mar '08

I am trying to compile Stunnel 4.21 on Solaris 10 with no luck. I am using Openssl 0.9.8f. After running configure, I run make and receive the following output: (truncated for size) ------------------------------------------------------- /bin/bash ../libtool --tag=CC --mode=link gcc -m64 -g -O2 -Wall -Wshadow -Wcast-align -Wpointer-arith -I/usr/local/ssl/include -o stunnel file.o client.o log.o options.o protocol.o network.o resolver.o ssl.o ctx.o verify.o sthreads.o stunnel.o pty.o -lz -ldl -lsocket -lnsl -lpthread -L/usr/local/ssl/lib -lssl -lcrypto gcc -m64 -g -O2 -Wall -Wshadow -Wcast-align -Wpointer-arith -I/usr/local/ssl/include -o stunnel file.o client.o log.o options.o protocol.o network.o resolver.o ssl.o ctx.o verify.o sthreads.o stunnel.o pty.o -lz -ldl -lsocket -lnsl -lpthread -L/usr/local/ssl/lib -lssl -lcrypto Undefined first referenced symbol in file BIO_set_flags protocol.o SSL_CTX_set_info_callback ctx.o ld: fatal: Symbol referencing errors. No output written to stunnel collect2: ld returned 1 exit status *** Error code 1 make: Fatal error: Command failed for target `stunnel' Current working directory /tmp/stunnel-4.20/src *** Error code 1 The following command caused the error: failcom='exit 1'; \ for f in x $MAKEFLAGS; do \ case $f in \ *=* | --[!k]*);; \ *k*) failcom='fail=yes';; \ esac; \ done; \ dot_seen=no; \ target=`echo all-recursive | sed s/-recursive//`; \ list='src doc tools'; for subdir in $list; do \ echo "Making $target in $subdir"; \ if test "$subdir" = "."; then \ dot_seen=yes; \ local_target="$target-am"; \ else \ local_target="$target"; \ fi; \ (cd $subdir && make $local_target) \ || eval $failcom; \ done; \ if test "$dot_seen" = "no"; then \ make "$target-am" || exit 1; \ fi; test -z "$fail" make: Fatal error: Command failed for target `all-recursive' I have tried Stunnel version 4.21, 4.20 with the same results. I have used Stunnel to help build several secure log servers in the past, and I've never had problems getting it compiled and installed. Any help would be greatly appreciated. cod3fr3ak

1 0

Re: [stunnel-users] stunnel not seeing ca bundle files
by Ludolf Holzheid 12 Mar '08

12 Mar '08

On Wed, 2008-03-12 09:05:17 -0700, sf.techguy(a)gmail.com wrote: > [..] > > Tried the command above on the 1st of 5 crt files (openssl pkcs12 -in > filename.crt -out filename.pem), and am getting this error: > > 19319:error:0D0680A8:asn1 encoding routines:ASN1_CHECK_TLEN:wrong > tag:tasn_dec.c:1007: > 19319:error:0D07803A:asn1 encoding routines:ASN1_ITEM_EX_D2I:nested asn1 > error:tasn_dec.c:305:Type=PKCS12 > > Any ideas why that might be happening? It seems your files do not contain a PKCS#12 structure (containing the key and the whole certificate chain). I should have recognize this by the fact you have one file for each certificate. Sorry. Maybe you should give the x509 openssl command a try: openssl x509 -in filename.crt -inform DES -out filename.pem HTH, Ludolf -- --------------------------------------------------------------- Ludolf Holzheid Tel: +49 621 339960 Bihl+Wiedemann GmbH Fax: +49 621 3392239 Floßwörthstraße 41 e-mail: lholzheid(a)bihl-wiedemann.de D-68199 Mannheim, Germany ---------------------------------------------------------------

1 0

stunnel not seeing ca bundle files
by sf.techguy＠gmail.com 12 Mar '08

12 Mar '08

I'm running stunnel 4.14 on Darwin 7.0.0 (Mac OS X Server 10.4.11), with OpenSSL 0.9.71. I purchased a security certificate from comodo, and was able to get it installed and working (mostly) with stunnel, but when I browse to the website, I get an error that the the browser can't verify the identity of the site, due to a problem w/the security cert. I called comodo, and they the error is caused by stunnel not seeing the intermediate certificate files that they told me need to be installed in order for their certificate to work. These files were: ComodoUTNServerCA.crt UTNAddTrustServerCA.crt AddTrustExternalCARoot.crt EssentialSSLCA.crt my_server_net.crt I modified stunnel.conf, and included this line: CApath = ca-certs/ (<- dir path, relative to chroot, containing above files) But the error persists. Can anyone tell me if what comodo support is telling me is even correct? I've verified that my .key and .crt files match by running: $ openssl x509 -noout -modulus -in server.crt | openssl md5 $ openssl rsa -noout -modulus -in server.key | openssl md5 And the output of both matches. Not sure what to try next. Any help much appreciated! Thanks in advance...

3 2

(no subject)
by Eddie Castelli 11 Mar '08

11 Mar '08

Dear all, Some years ago I have installed Apache, PHP and mySQL. Now I need to update. Unfortunately I can't remember if I installed a package or individual program elements. Could please someone help me and guide me trough as I need urgently the update. Many thanks in advance. -- best regards Eddie

3 3

behaviour on remote disconnection
by Marek Jarycki 11 Mar '08

11 Mar '08

Hi. I'm using 3.26 version of stunnel. How to disallow creating local sockets (on client side) when remote (server) is not available? My application detects socket creation errors and it works without stunnel, but in stunnel the socket is always opening correctly so I have to wait for timeouts to discover network disconnection. Regards, deimoss

2 1

Re: [stunnel-users] unknown protocol error (140770FC) with xinetd
by Ian Pilcher 10 Mar '08

10 Mar '08

Ian Pilcher wrote: > ... and the (non-working) server-side stunnel configuration: > > cert = /etc/stunnel/rsyncs_cert.pem > key = /etc/stunnel/private/rsyncs_key.pem > client = no > pid = /var/run/stunnel.pid > RNDfile = /var/run/stunnel/random_seed > foreground = yes > debug = debug > output = /var/run/stunnel/stunnel.log > > exec = /usr/bin/rsync > execargs = rsync --daemon Amazing how sending off the question gets the old brain cells working (or maybe it was the second cup of coffee). The problem was "foreground = yes". Since xinet.d redirects stderr back over the network connection, the stunnel startup messages were going back to the client. For some reason, the client didn't think that was a proper SSL handshake. ;-) -- ======================================================================== Ian Pilcher arequipeno(a)gmail.com ========================================================================

1 0

unknown protocol error (140770FC) with xinetd
by Ian Pilcher 10 Mar '08

10 Mar '08

I am trying to set up rsync over SSL with stunnel. I've got things working when I run stunnel standalone on both systems, but I have been unable to get things working when I run stunnel via xinetd on the server (haven't tried using xinet on the client yet). Here is the (standalone) server configuration that works: cert = /etc/stunnel/rsyncs_cert.pem key = /etc/stunnel/private/rsyncs_key.pem client = no pid = /var/run/stunnel.pid RNDfile = /var/run/stunnel/random_seed foreground = yes debug = debug ;output = /var/run/stunnel/stunnel.log [rsyncs] accept = 2873 exec = /usr/bin/rsync execargs = rsync --daemon Here is the (non-working) server-side xinetd configuration: # default: off # description: rsync tunneled over SSL (stunnel) service rsyncs { disable = no flags = IPv4 socket_type = stream wait = no user = root server = /usr/sbin/stunnel } ... and the (non-working) server-side stunnel configuration: cert = /etc/stunnel/rsyncs_cert.pem key = /etc/stunnel/private/rsyncs_key.pem client = no pid = /var/run/stunnel.pid RNDfile = /var/run/stunnel/random_seed foreground = yes debug = debug output = /var/run/stunnel/stunnel.log exec = /usr/bin/rsync execargs = rsync --daemon Trying to connect to the non-working configuration with openssl s_client, I get: CONNECTED(00000003) write to 0x9b9e938 [0x9b9f068] (133 bytes => 133 (0x85)) 0000 - 80 83 01 03 01 00 5a 00-00 00 20 00 00 39 00 00 ......Z... ..9.. 0010 - 38 00 00 35 00 00 16 00-00 13 00 00 0a 07 00 c0 8..5............ 0020 - 00 00 33 00 00 32 00 00-2f 03 00 80 00 00 66 00 ..3..2../.....f. 0030 - 00 05 00 00 04 01 00 80-00 00 63 00 00 62 00 00 ..........c..b.. 0040 - 15 00 00 12 00 00 09 06-00 40 00 00 65 00 00 64 .........@..e..d 0050 - 00 00 14 00 00 11 00 00-08 00 00 06 04 00 80 00 ................ 0060 - 00 03 02 00 80 3e b4 89-34 c1 e6 93 97 2d 8d ca .....>..4....-.. 0070 - 90 e1 4c 51 e1 71 c7 54-ed 52 0c 6e 2e 47 90 9b ..LQ.q.T.R.n.G.. 0080 - 8e c6 73 3d d8 ..s=. read from 0x9b9e938 [0x9ba45c8] (7 bytes => 7 (0x7)) 0000 - 32 30 30 38 2e 30 33 2008.03 18523:error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol:s23_clnt.c:567: The server log shows: 2008.03.10 11:32:23 LOG5[18563:3086215456]: stunnel 4.20 on i386-redhat-linux-gnu with OpenSSL 0.9.8b 04 May 2006 2008.03.10 11:32:23 LOG5[18563:3086215456]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP 2008.03.10 11:32:23 LOG7[18563:3086215456]: stunnel started 2008.03.10 11:32:23 LOG7[18563:3086215456]: FD 0 in non-blocking mode 2008.03.10 11:32:23 LOG7[18563:3086215456]: FD 1 in non-blocking mode 2008.03.10 11:32:23 LOG7[18563:3086215456]: FD 4 in non-blocking mode 2008.03.10 11:32:23 LOG7[18563:3086215456]: FD 5 in non-blocking mode 2008.03.10 11:32:23 LOG7[18563:3086215456]: Connection from 127.0.0.1:38481 permitted by libwrap 2008.03.10 11:32:23 LOG5[18563:3086215456]: stunnel accepted connection from 127.0.0.1:38481 2008.03.10 11:32:23 LOG7[18563:3086215456]: SSL state (accept): before/accept initialization 2008.03.10 11:32:23 LOG7[18563:3086215456]: SSL state (accept): SSLv3 read client hello A 2008.03.10 11:32:23 LOG7[18563:3086215456]: SSL state (accept): SSLv3 write server hello A 2008.03.10 11:32:23 LOG7[18563:3086215456]: SSL state (accept): SSLv3 write certificate A 2008.03.10 11:32:23 LOG7[18563:3086215456]: SSL state (accept): SSLv3 write server done A 2008.03.10 11:32:23 LOG3[18563:3086215456]: SSL_accept: Peer suddenly disconnected 2008.03.10 11:32:23 LOG5[18563:3086215456]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.03.10 11:32:23 LOG7[18563:3086215456]: stunnel finished (0 left) Anyone have any ideas? Thanks! -- ======================================================================== Ian Pilcher arequipeno(a)gmail.com ========================================================================

1 0

Windows FIPS compile
by Joe Kemp 10 Mar '08

10 Mar '08

I am trying to get stunnel to compile against a fips openssl. I read in install.fips that it is not yet supported. Is anyone currently working on this? If so I would love to join forces or any info on the specific issues would help get me started. Thanks. -Joe

2 2

help
by Paque Alexandre 10 Mar '08

10 Mar '08

Hi, I am writing you because I haven’t found solution on the web. In order to secure a connection between servers (1 client and 1 server stunnel), I chose the stunnel tool. However, I have an issue when I unplugged the network cable. If I unplug the network cable on the remote side, Stunnel will detect the cutting only on the local side. In fact, I would like to configure stunnel client in order to detect the unplugged network cable on server side. I tried many configurations but without success: socket = l:TCP_NODELAY=1 socket = l:SO_LINGER=1:1 socket = l:SO_KEEPALIVE=1 socket = l:SO_REUSEADDR=1 socket = r:TCP_NODELAY=1 socket = r:SO_LINGER=1:1 socket = r:SO_KEEPALIVE=1 socket = r:SO_REUSEADDR=1 Could you help me? Thanks you in advance, Alex _____________________________________________________________________________ Envoyez avec Yahoo! Mail. Une boite mail plus intelligente http://mail.yahoo.fr

1 0

ssl mutual authentication
by Simon Santoro 09 Mar '08

09 Mar '08

Hello all. I am sure I am missing something really obvious here, but I can't get this to work. HTTP client (localhost) -> stunnel (localhost) -> HTTPS server (remote host) I'd like to user stunnel as proxy for my HTTP requests that go to a HTTPS server (apache if that makes any difference) on a remote host (fresh) that uses a cert to authenticate the request. I wrote this little ruby script[2] to test the request, and it works, so i am sure that on the server/client/cert end there is no problem. Now I'd like to use a simple wget or w3m to get the file from the remote server trough stunnel, but I keep getting SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed The only difference between the ruby script and the stunnel setup is that in the ruby script i have two different files for the crt and the key, and with stunnel I joined those files together with cat s2(a)magnesium.key s2(a)magnesium.crt >s2(a)magnesium.pam on localhost i run: $ stunnel -r fresh:443 -d localhost:11443 -c -f -D 7 -c -S 0 -p certs/s2(a)magnesium.pem -a certs/ -A certs/freshCA.crt -v 3 -P none 2008.03.09 21:21:38 LOG5[9026:3082811056]: Using 'fresh.443' as tcpwrapper service name 2008.03.09 21:21:38 LOG7[9026:3082811056]: RAND_status claims sufficient entropy for the PRNG 2008.03.09 21:21:38 LOG6[9026:3082811056]: PRNG seeded successfully 2008.03.09 21:21:38 LOG7[9026:3082811056]: Certificate: certs/s2(a)magnesium.pem Enter PEM pass phrase: 2008.03.09 21:21:40 LOG7[9026:3082811056]: cert_defaults is 0 2008.03.09 21:21:40 LOG7[9026:3082811056]: cert_dir is certs/ 2008.03.09 21:21:40 LOG7[9026:3082811056]: cert_file is certs/freshCA.crt 2008.03.09 21:21:40 LOG7[9026:3082811056]: Loaded verify certificates from certs/freshCA.crt 2008.03.09 21:21:40 LOG7[9026:3082811056]: Set verify directory to certs/ 2008.03.09 21:21:40 LOG5[9026:3082811056]: Peer certificate location certs/ 2008.03.09 21:21:40 LOG5[9026:3082811056]: stunnel 3.26 on i486-pc-linux-gnu PTHREAD+LIBWRAP with OpenSSL 0.9.8e 23 Feb 2007 2008.03.09 21:21:40 LOG7[9026:3082811056]: No pid file being created 2008.03.09 21:21:40 LOG5[9026:3082811056]: FD_SETSIZE=1024, file ulimit=1024 -> 500 clients allowed 2008.03.09 21:21:40 LOG7[9026:3082811056]: SO_REUSEADDR option set on accept socket 2008.03.09 21:21:40 LOG7[9026:3082811056]: fresh.443 bound to 127.0.0.1:11443 then, I try to connect to localhost:11443 with w3m, in the hope that my HTTP request is encrypted and send to fresh:443, and stunnel should check if the PEER certificate is correct, but all I get is: 2008.03.09 21:26:44 LOG7[9214:3082585776]: fresh.443 accepted FD=6 from 127.0.0.1:43888 2008.03.09 21:26:44 LOG7[9214:3082423184]: fresh.443 started 2008.03.09 21:26:44 LOG5[9214:3082423184]: fresh.443 connected from 127.0.0.1:43888 2008.03.09 21:26:44 LOG7[9214:3082423184]: fresh.443 connecting 192.168.0.2:443 2008.03.09 21:26:44 LOG7[9214:3082423184]: Remote FD=7 initialized 2008.03.09 21:26:44 LOG7[9214:3082423184]: Relying on OpenSSL RSA Blinding. 2008.03.09 21:26:44 LOG7[9214:3082423184]: SSL state (connect): before/connect initialization 2008.03.09 21:26:44 LOG7[9214:3082423184]: SSL state (connect): SSLv3 write client hello A 2008.03.09 21:26:45 LOG7[9214:3082423184]: SSL state (connect): SSLv3 read server hello A 2008.03.09 21:26:45 LOG5[9214:3082423184]: VERIFY OK: depth=1, /C=IT/O=S2 corp./CN=S2/emailAddress=s2@fresh 2008.03.09 21:26:45 LOG4[9214:3082423184]: VERIFY ERROR ONLY MY: no cert for /C=IT/O=S2 corp./CN=fresh 2008.03.09 21:26:45 LOG7[9214:3082423184]: SSL alert (write): fatal: certificate unknown 2008.03.09 21:26:45 LOG3[9214:3082423184]: SSL_connect: error:14090086:SSL routines:SSL3_GET_SERVER_CERTIFICATE:certificate verify failed 2008.03.09 21:26:45 LOG7[9214:3082423184]: fresh.443 finished (0 left) Did I something wrong? What am I missing? Please help me, it would save me really a lot of work on the client side if I could get this going. Thanks! Simon [1] ---------------------------------------------------- #!/usr/bin/ruby SERVER_CERT_FILE = "certs/freshCA.crt" CERT_FILE = "certs/s2(a)magnesium.crt" CERT_FILE_KEY = "certs/s2(a)magnesium.key" require 'net/https' https = Net::HTTP.new('fresh', 443) https.use_ssl = true #client certificates https.cert = OpenSSL::X509::Certificate.new( File.read(CERT_FILE) ) https.key = OpenSSL::PKey::RSA.new( File.read(CERT_FILE_KEY), 'panza') #server certificate https.ca_file = SERVER_CERT_FILE https.verify_mode = OpenSSL::SSL::VERIFY_PEER #VERIFY_NONE https.read_timeout = 120 https.start do |https| request = Net::HTTP::Get.new('/notes/test_auth') # request.basic_auth 's', 'x' response = https.request(request) response.value puts response.body end ----------------------------------------------------

1 1

Connection Problem on Vista
by Earl Stenlund 08 Mar '08

08 Mar '08

I setup stunnel this afternoon to work with Newshosting. It was working fine but now I cannot connect using my desktop running Vista. I receive this message: SSL_connect: 14094406: error:14094406:SSL routines:SSL3_READ_BYTES:sslv3 alert decompression failure Stunnel connects fine on my laptop which is also running Vista. Help, Earl Stenlund

1 0

Username authentication
by Joe Kemp 07 Mar '08

07 Mar '08

I want to try to use stunnel as a "simple" client vpn. It solves all of my encryption issues but I would like to verify a username/password before it lets the traffic through. I didn't see any patches or hacks out there that did this. Has this been attempted before or am I on my own. I would also be interested in other solutions based on openssl that are not network device level VPNs clients. Already using client side certificates and I know that is the normal SSL authentication mechanism.... Thanks. -Joe

2 2

Send delay?
by Colin E 04 Mar '08

04 Mar '08

Please can you help a hardware engineer, who almost understands what he's doing. I'm using sTunnel 4.21 on a Windows 2000 machine, to provide a TCP program with SSL. This works very well. However the application is very time critical and must finish sending block of XML as quickly as possible. I suspect there is a delay in either sTunnel or openSSL, while it waits to see if there are any more characters before sending a packet. Is this correct? If it is, is there a work round. -- View this message in context: http://www.nabble.com/Send-delay--tp15826043p15826043.html Sent from the Stunnel - Users mailing list archive at Nabble.com.

2 4

Newbie
by Rod Bodden 03 Mar '08

03 Mar '08

Hello to all! I'm trying to set up Stunnel for the first time. I hope to use it with HFS. I followed directions found on HFS website and ran into a problem creating a pem file. Could someone direct me to information about installing Stunnel? Or help me directly by sharing with me how to install Stunnel? I sincerely appreciate your help!!! Rod

1 0

Bugs
by Jan.Haering＠rz-kiru.de 29 Feb '08

29 Feb '08

Hallo, I have found 2 Bugs. 1. In Version 4.21: No Taskbar Icon is displayed when the service starts. Although the taskbar ist set in config file (taskbar = yes). 2. In probably all Versions (I have testet 4.18, 4.20 and 4.21) If i started the service over Start -> Programm Files -> stunnel -> Service Start (stunnel -start) no log files are written. If i start the service over services.msc from Windows or "sc start stunnel" log files are written. The icon log displayed the following message in version 4.18 and 4.20 2007.11.09 09:41:03 LOG3[2396:2400]: stunnel.log: Broken pipe (32) 2007.11.09 09:41:03 LOG3[2396:2400]: Unable to open output file: stunnel.log Best Regards Jan

5 9

webdav with stunnel and 'net use'
by detlef oertel 26 Feb '08

26 Feb '08

Hello list, what we want to do, is mounting a webdav share via https on a windows xp box. Therefore we tried to tunnel localhost:80 to server:443 The name of the test server is here bonifax. What works is: mounting direct via http net use x: "http://bonfax/opsi-depot" mounting over stunnel with 'Add Network Place Wizard' or RunDLL32.EXE netplwiz.dll,AddNetPlaceRunDll using the address: http://localhost/opsi-depot What fails is: ---------------------------- C:\WINDOWS>net use y: "http://localhost/opsi-depot" Systemfehler 67 aufgetreten. Der Netzwerkname wurde nicht gefunden. -------- Error 67 'network name not found' ------------------------ part of stunnel.conf ------------------------ ; Some debugging stuff useful for troubleshooting debug = 7 output = stunnel.log ; Use it for client mode client = yes ; Service-level configuration verify=0 [psuedo-https] accept = 80 connect = bonifax:443 TIMEOUTclose = 0 ------------------------ This is the log: ----------------------------- 2008.02.26 20:04:11 LOG7[1560:1404]: psuedo-https accepted FD=176 from 127.0.0.1:1861 2008.02.26 20:04:11 LOG7[1560:1404]: Creating a new thread 2008.02.26 20:04:11 LOG7[1560:1404]: New thread created 2008.02.26 20:04:11 LOG7[1560:3308]: psuedo-https started 2008.02.26 20:04:11 LOG7[1560:3308]: FD 176 in non-blocking mode 2008.02.26 20:04:11 LOG7[1560:3308]: TCP_NODELAY option set on local socket 2008.02.26 20:04:11 LOG5[1560:3308]: psuedo-https accepted connection from 127.0.0.1:1861 2008.02.26 20:04:11 LOG7[1560:3308]: FD 208 in non-blocking mode 2008.02.26 20:04:11 LOG7[1560:3308]: psuedo-https connecting 192.168.1.14:443 2008.02.26 20:04:11 LOG7[1560:3308]: connect_wait: waiting 10 seconds 2008.02.26 20:04:11 LOG7[1560:3308]: connect_wait: connected 2008.02.26 20:04:11 LOG5[1560:3308]: psuedo-https connected remote server from 192.168.2.234:1862 2008.02.26 20:04:11 LOG7[1560:3308]: Remote FD=208 initialized 2008.02.26 20:04:11 LOG7[1560:3308]: TCP_NODELAY option set on remote socket 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): before/connect initialization 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 write client hello A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 read server hello A 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY IGNORE: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY OK: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY IGNORE: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY OK: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY IGNORE: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG5[1560:3308]: VERIFY OK: depth=0, /C=DE/ST=Rheinland-Pfalz/O=uib/CN=bonifax.uib.local 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 read server certificate A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 read server key exchange A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 read server done A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 write client key exchange A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 write change cipher spec A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 write finished A 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 flush data 2008.02.26 20:04:11 LOG7[1560:3308]: SSL state (connect): SSLv3 read finished A 2008.02.26 20:04:11 LOG7[1560:3308]: 1 items in the session cache 2008.02.26 20:04:11 LOG7[1560:3308]: 1 client connects (SSL_connect()) 2008.02.26 20:04:11 LOG7[1560:3308]: 1 client connects that finished 2008.02.26 20:04:11 LOG7[1560:3308]: 0 client renegotiations requested 2008.02.26 20:04:11 LOG7[1560:3308]: 0 server connects (SSL_accept()) 2008.02.26 20:04:11 LOG7[1560:3308]: 0 server connects that finished 2008.02.26 20:04:11 LOG7[1560:3308]: 0 server renegotiations requested 2008.02.26 20:04:11 LOG7[1560:3308]: 0 session cache hits 2008.02.26 20:04:11 LOG7[1560:3308]: 0 session cache misses 2008.02.26 20:04:11 LOG7[1560:3308]: 0 session cache timeouts 2008.02.26 20:04:11 LOG6[1560:3308]: SSL connected: new session negotiated 2008.02.26 20:04:11 LOG6[1560:3308]: Negotiated ciphers: DHE-RSA-AES256-SHA SSLv3 Kx=DH Au=RSA Enc=AES(256) Mac=SHA1 2008.02.26 20:04:26 LOG7[1560:3308]: SSL alert (read): warning: close notify 2008.02.26 20:04:26 LOG7[1560:3308]: SSL closed on SSL_read 2008.02.26 20:04:26 LOG7[1560:3308]: Socket write shutdown 2008.02.26 20:04:26 LOG7[1560:3308]: SSL write shutdown 2008.02.26 20:04:26 LOG7[1560:3308]: SSL alert (write): warning: close notify 2008.02.26 20:04:26 LOG6[1560:3308]: SSL_shutdown successfully sent close_notify 2008.02.26 20:04:26 LOG5[1560:3308]: Connection closed: 311 bytes sent to SSL, 971 bytes sent to socket 2008.02.26 20:04:26 LOG7[1560:3308]: psuedo-https finished (0 left) ---------------------- stunnel version 4.21 Anyone any ideas what went wrong and how to go around ? regards detlef -- detlef oertel Tel. 06131 / 27561-10 Fax 06131 / 27561-22 eMail: d.oertel(a)uib.de uib umwelt informatik büro gmbh Firmensitz / Anschrift: Bonifaziusplatz 1B 55118 Mainz internet: www.uib.de Geschäftsführer: dr. detlef oertel Handelsregister: Amtsgericht Mainz HRB 6942

1 0

Sun Niagra2
by Steve Harris 25 Feb '08

25 Feb '08

Hi all, Has built stunnel on the Sun Niagra 2 platform to take advantage of the on-chip crypto? Cheers - Steve -- NOTE: This message is confidential and may not be disclosed to anyone without the express written consent of the sender. ********************************************************************* PH: 250-381-1335 xt125

1 0

Log & Timestamp question
by Edward Blake 25 Feb '08

25 Feb '08

Stunnel currently logs times in the format hh:mm:ss. Is there a way to get a more exact time, such as UTC or down to the millisecond? Edward Blake

1 0

VNC, STUNNEL & HTTPTUNNEL ----- (((VNC)SSL)HTTP)
by sean bhola 25 Feb '08

25 Feb '08

I need to connect to my PC at work from home. The firewall at work has only ports 80 (http) and 443 (https) open, I also think there is a proxy. I was wondering which scenario would work. 1: To encapsulate vnc traffic within ssl using stunnel and pass it though port 443, OR 2: To encapsulate vnc traffic within ssl using stunnel, then encapsulate that within http using httptunnel and pass through port 80 OR 3: To encapsulate vnc traffic within http, then encapsulate that within ssl using stunnel and pass through port 443 I read this article http://www.securityfocus.com/infocus/1677 but still had the concerns above. Help would be much appreciated. Thanks. Sean

2 1

running SWAT through stunnel and uid/gid question
by Wayne Sherman 24 Feb '08

24 Feb '08

(using STunnel 4.21 on gentoo) I can successfully run the samba SWAT tool through stunnel with these settings: [swat] exec = /usr/sbin/swat execargs = swat accept = 901 But, I have discovered that in order to successfully log-in to swat, I have to comment these lines in the global section of stunnel.conf: #setuid = stunnel #setgid = stunnel Is there a way to specify the UID/GID per service (like the xinetd service level option "user") instead of globally? Thanks, Wayne

1 0

Re: [stunnel-users] Stunnel Configuration
by paque_alexandre＠yahoo.fr 19 Feb '08

19 Feb '08

1 0

Web based CRL file
by Edouard Dessioux 18 Feb '08

18 Feb '08

Hi, I have several questions about the CRL file : - is the file reloaded before each connection or is it first loaded at the stunnel startup ? - do you have any idea of how to use a web-based CRL file ? Regards, Edouard DESSIOUX Directeur de Projets Tibco Mobile 3, rue Danton - 92240 Malakoff Tél : +33 (0)1 55 58 04 59 - Fax : +33 (0)1 55 58 03 89 - Mob. +33 (0)6 34 02 61 54 E-mail : edessioux(a)tibco.fr - www.tibcomobile.fr Faites un geste pour la planète, n'imprimez ce message que si nécessaire.

1 0

Installing Stunnel 4.21 on Shared Hosting Server with OpenSSL 0.9.6g
by Norbert Hoeller 17 Feb '08

17 Feb '08

I am trying to install Stunnel 4.21 on a 1&1 shared Linux server. PHP Settings claims that OpenSSL 0.9.6g is installed. I unpacked the tar file and ran "./configure" using --prefix to override the /usr/local/ install directory. I had one warning: "configure: WARNING: Openssl engine header not found". In terms of threading model, I got "configure: PTHREAD thread model detected". When I tried to run "make", I got "common.h:318: #error OpenSSL library compiled without thread support". Is there a way around this? I tried running "./configure" with --with-threads=ucontext/fork, but that gave a bunch of "OCSP_" undeclared variables and "Error 1". Thanks, Norbert

2 2

Validation of certificate
by Edouard Dessioux 12 Feb '08

12 Feb '08

Hi everyone, I'm new to this mailing list, but before my asking question, i checked the archive but did not find anything related, so, it may be out of topic. Well, for the moment, I can't provide all the informations asked like « output of ... command », but I'm not sure this is relevant for my question. The informations I can provide are : * I use stunnel to terminate a tls connection between a mobile phone and a web application server. * I use stunnel to validate my clients certificates. * The stunnel will be running on a windows 2003 server * I guess I would be using the last available version of stunnel * The certificates would be generated by a PKI different from openssl What I would like to do is : I have a PKI already deployed with many certificates, and I want only a small subset of them to be validated by the stunnel. I thought of using multiples OU fields to « tags » some users, but can this fields be checked by stunnel ? Do you guys have any feedback on this or any alternative way to do something equivalent ? Regards, Edouard DESSIOUX Directeur de Projets Tibco Mobile 3, rue Danton - 92240 Malakoff Tél : +33 (0)1 55 58 04 59 - Fax : +33 (0)1 55 58 03 89 - Mob. +33 (0)6 34 02 61 54 E-mail : edessioux(a)tibco.fr - www.tibcomobile.fr Faites un geste pour la planète, n'imprimez ce message que si nécessaire.

1 0

signal 15 terminating stunnel on start in Ubuntu
by Rick Barry 11 Feb '08

11 Feb '08

I am trying to use Stunnel 4.2 on Ubuntu 7.1. When I use this command to start Stunnel: /etc/init.d/stunnel4 start I get this output: Starting SSL tunnels: 2008.02.11 09:00:05 LOG7[6780:3082872496]: RAND_status claims sufficient entropy for the PRNG 2008.02.11 09:00:05 LOG7[6780:3082872496]: PRNG seeded successfully 2008.02.11 09:00:05 LOG7[6780:3082872496]: SSL context initialized for service nntp [Failed: /etc/stunnel/stunnel.conf] You should check that you have specified the pid= in you configuration file In the log entry @ /var/log/stunnel4/stunnel.log: 2008.02.11 08:33:54 LOG5[5468:3082974896]: stunnel 4.20 on i486-pc-linux- gnu with OpenSSL 0.9.8e 23 Feb 2007 2008.02.11 08:33:54 LOG5[5468:3082974896]: ThreadingTHREAD SSL:ENGINE SocketsOLL,IPv6 Auth:LIBWRAP 2008.02.11 08:33:54 LOG6[5468:3082974896]: file ulimit = 1024 (can be changed with 'ulimit -n') 2008.02.11 08:33:54 LOG6[5468:3082974896]: poll() used - no FD_SETSIZE limit for file descriptors 2008.02.11 08:33:54 LOG5[5468:3082974896]: 500 clients allowed 2008.02.11 08:33:54 LOG7[5468:3082974896]: FD 4 in non-blocking mode 2008.02.11 08:33:54 LOG7[5468:3082974896]: FD 5 in non-blocking mode 2008.02.11 08:33:54 LOG7[5468:3082974896]: FD 6 in non-blocking mode 2008.02.11 08:33:54 LOG7[5468:3082974896]: SO_REUSEADDR option set on accept socket 2008.02.11 08:33:54 LOG7[5468:3082974896]: nntp bound to 127.0.0.1:119 2008.02.11 08:33:54 LOG7[5517:3082974896]: Created pid file /stunnel4.pid 2008.02.11 08:39:10 LOG5[5517:3082974896]: Received signal 15; terminating 2008.02.11 08:39:10 LOG7[5517:3082974896]: removing pid file /stunnel4.pid Signal 15 seems to be the linchpin here, but my research hasn't brought up anything conclusive. Any pointers? Thank you. Rick Barry ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping

1 0

Stunnel Problem - 935 Maximum connections on Windows
by Iain ＠ Mobile777 11 Feb '08

11 Feb '08

We are having some difficulties when using Stunnel under load. The maximum sessions running on a Windows servers seems to be set at 935 concurrent connections. Our system uses Stunnel to route incoming SSL traffic on port 443 through Openssl to a Java application on port 26789. Both Stunnel and the Java server run on a Windows 2003 server, with between 4GB and 8GB of ram, both running Dual Intel Zeon 2Ghz processors. The problem is that when the Stunnel server is under load, the maximum connections that we can achieve is 935 concurrent connections (Windows threads between 936 and 938), before Stunnel throws errors and starts closing the connections. Here is some additional information about our system: Stunnel version 4.21 (latest Windows binaries) OpenSSL 0.9.8g Stunnel Config: ----------------------------------------------------------------------- debug = 7 output = stunnel.log # Use it for client mode #client = yes #TIMEOUTidle = 60 socket = a:TCP_NODELAY=1 socket = a:SO_LINGER=a:0 socket = r:TCP_NODELAY=1 socket = r:SO_LINGER=r:0 socket = l:TCP_NODELAY=1 socket = l:SO_LINGER=l:0 # Service-level configuration [JAVASERVER] #session = 5000 accept = 443 connect = localhost: 26789 # delay = yes #transparent = yes ----------------------------------------------------------------------- We have tried various options, such as the TIMEOUTidle and session parameters for Stunnel, but with no luck. We have even tried running Stunnel separately on a Linux distro (FC4), with even less positive results. The pertinent errors in the log files are as follows: Connection closed: 178664 bytes sent to SSL, 44646 bytes sent to socket SSL alert (write): warning: close notify _beginthread: Exec format error (8) Connection rejected: create_client failed I can mail full logs of our load tests as well. Our server will be expected to handle upwards of 10000 connections at a time, we've been researching for about a week now how to get the performance we need, but so far nothing. Iain

3 3

Stunnel + Hybrid-ircd issues
by Strykar 09 Feb '08

09 Feb '08

Hi, I'm running Hybrid-7.2 on two PCs, one of which has a dynamic IP, as Hybrid wants a static IP and not hostname in its connect section, we are trying to use stunnel to encrypt server<->server communication. Server 1 specs (To which I'm connecting) running FreeBSD 6.3: *********************************************** #stunnel -version stunnel 4.05 on amd64-unknown-freebsd5.3 PTHREAD+LIBWRAP with OpenSSL 0.9.7e 25 Oct 2004 Global options cert = /usr/local/etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH debug = 5 key = /usr/local/etc/stunnel/stunnel.pem pid = /var/tmp/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes #gcc -v Using built-in specs. Configured with: FreeBSD/amd64 system compiler Thread model: posix gcc version 3.4.6 [FreeBSD] 20060305 #uname FreeBSD 6.3-STABLE FreeBSD 6.3-STABLE #6: Tue Jan 22 13:23:51 GMT 2008 root@:/usr/obj/usr/src/sys/SVR1 amd64 Server 2 specs (From which I'm connecting via stunnel as a client) running OpenBSD 4.2: ****************************************************************** #stunnel -version stunnel 4.20 on i386-unknown-openbsd4.2 with OpenSSL 0.9.7j 04 May 2006 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /usr/local/var/run/stunnel/stunnel.pid RNDbytes = 64 RNDfile = /dev/arandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none # gcc -v Reading specs from /usr/lib/gcc-lib/i386-unknown-openbsd4.2/3.3.5/specs Configured with: Thread model: single gcc version 3.3.5 (propolice) # cat stunnel.conf: cert = /etc/ssl/private/stunnel.pem key = /etc/ssl/private/rsa.key setuid = _stunnel setgid = _stunnel pid = /var/run/stunnel.pid socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 debug = 7 foreground = yes [irc] client = yes accept = localhost:994 connect = xxx.xxx.xxx.xxx:994 Here's the debug logged to stderr: # stunnel 2008.02.08 19:34:54 LOG7[11904:2237644800]: Snagged 64 random bytes from /dev/arandom 2008.02.08 19:34:54 LOG7[11904:2237644800]: RAND_status claims sufficient entropy for the PRNG 2008.02.08 19:34:54 LOG7[11904:2237644800]: PRNG seeded successfully 2008.02.08 19:34:54 LOG7[11904:2237644800]: Certificate: /etc/ssl/private/stunnel.pem 2008.02.08 19:34:54 LOG7[11904:2237644800]: Certificate loaded 2008.02.08 19:34:54 LOG7[11904:2237644800]: Key file: /etc/ssl/private/rsa.key 2008.02.08 19:34:54 LOG7[11904:2237644800]: Private key loaded 2008.02.08 19:34:54 LOG7[11904:2237644800]: SSL context initialized for service irc 2008.02.08 19:34:54 LOG5[11904:2237644800]: stunnel 4.20 on i386-unknown-openbsd4.2 with OpenSSL 0.9.7j 04 May 2006 2008.02.08 19:34:54 LOG5[11904:2237644800]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP 2008.02.08 19:34:54 LOG6[11904:2237644800]: file ulimit = 128 (can be changed with 'ulimit -n') 2008.02.08 19:34:54 LOG6[11904:2237644800]: poll() used - no FD_SETSIZE limit for file descriptors 2008.02.08 19:34:54 LOG5[11904:2237644800]: 61 clients allowed 2008.02.08 19:34:54 LOG7[11904:2237644800]: FD 6 in non-blocking mode 2008.02.08 19:34:54 LOG7[11904:2237644800]: FD 7 in non-blocking mode 2008.02.08 19:34:54 LOG7[11904:2237644800]: FD 8 in non-blocking mode 2008.02.08 19:34:54 LOG7[11904:2237644800]: SO_REUSEADDR option set on accept socket 2008.02.08 19:34:54 LOG7[11904:2237644800]: irc bound to 127.0.0.1:994 2008.02.08 19:34:54 LOG7[11904:2237644800]: Created pid file /var/run/stunnel.pid 2008.02.08 19:35:15 LOG7[11904:2237644800]: irc accepted FD=9 from 127.0.0.1:8579 2008.02.08 19:35:15 LOG7[11904:2336256000]: irc started 2008.02.08 19:35:15 LOG7[11904:2336256000]: FD 9 in non-blocking mode 2008.02.08 19:35:15 LOG7[11904:2336256000]: TCP_NODELAY option set on local socket 2008.02.08 19:35:15 LOG7[11904:2336256000]: FD 10 in non-blocking mode 2008.02.08 19:35:15 LOG7[11904:2336256000]: FD 11 in non-blocking mode 2008.02.08 19:35:15 LOG7[11904:2336256000]: Connection from 127.0.0.1:8579 permitted by libwrap 2008.02.08 19:35:15 LOG5[11904:2336256000]: irc accepted connection from 127.0.0.1:8579 2008.02.08 19:35:15 LOG7[11904:2336256000]: FD 10 in non-blocking mode 2008.02.08 19:35:15 LOG7[11904:2336256000]: irc connecting 69.50.175.50:994 2008.02.08 19:35:15 LOG7[11904:2336256000]: connect_wait: waiting 10 seconds 2008.02.08 19:35:15 LOG7[11904:2237644800]: Cleaning up the signal pipe 2008.02.08 19:35:15 LOG6[11904:2237644800]: Child process 26562 finished with code 0 2008.02.08 19:35:15 LOG7[11904:2336256000]: connect_wait: connected 2008.02.08 19:35:15 LOG5[11904:2336256000]: irc connected remote server from 192.168.1.101:42954 2008.02.08 19:35:15 LOG7[11904:2336256000]: Remote FD=10 initialized 2008.02.08 19:35:15 LOG7[11904:2336256000]: TCP_NODELAY option set on remote socket 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): before/connect initialization 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 write client hello A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 read server hello A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 read server certificate A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 read server done A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 write client key exchange A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 write change cipher spec A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 write finished A 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 flush data 2008.02.08 19:35:15 LOG7[11904:2336256000]: SSL state (connect): SSLv3 read finished A 2008.02.08 19:35:15 LOG7[11904:2336256000]: 1 items in the session cache 2008.02.08 19:35:15 LOG7[11904:2336256000]: 1 client connects (SSL_connect()) 2008.02.08 19:35:15 LOG7[11904:2336256000]: 1 client connects that finished 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 client renegotiations requested 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 server connects (SSL_accept()) 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 server connects that finished 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 server renegotiations requested 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 session cache hits 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 session cache misses 2008.02.08 19:35:15 LOG7[11904:2336256000]: 0 session cache timeouts 2008.02.08 19:35:15 LOG6[11904:2336256000]: SSL connected: new session negotiated 2008.02.08 19:35:15 LOG6[11904:2336256000]: Negotiated ciphers: AES256-SHA SSLv3 Kx=RSA Au=RSA Enc=AES(256) Mac=SHA1 2008.02.08 19:35:15 LOG3[11904:2336256000]: SSL_read: Connection reset by peer (54) 2008.02.08 19:35:15 LOG5[11904:2336256000]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.02.08 19:35:15 LOG7[11904:2336256000]: irc finished (0 left) What is going on here with "SSL_read: Connection reset by peer (54)"? This process keeps repeating itself without the ircd's linking. - S

1 0

Verify Question
by Vlad 08 Feb '08

08 Feb '08

I'm sure that people have asked this question before, but i cannot find a way to search the archives. What is the difference between verify = 2 and verify = 3, The docs indicate that with verify = 3it will check against locally installed certs, but what does that mean, if I use =2 then what does it verify the cert against? Also a second question: I'm pretty sure that the answer is no but is it possible to do the following, have a cert A that is signed by CA cert B which itself is singed by a self-signed cert C. If the clients has C installed as a trusted cert, and it received A, is there anyway for it to validate A. I realize that these are probably pretty novice questions so I really appreciate your patience, thanks!

1 0

xforwardedfor in config
by Jill Rochelle 04 Feb '08

04 Feb '08

I am new to stunnel and the ssl world. I'm trying to get stunnel to work with haproxy for load balancing https requests. In the haproxy documentation it says that the stunnel.conf should look like this cert=/etc/stunnel/stunnel.pem setuid=stunnel setgid=proxy socket=l:TCP_NODELAY=1 socket=r:TCP_NODELAY=1 [https] accept=192.168.1.1:443 connect=192.168.1.1:80 xforwardedfor=yes However, when I try to start the stunnel service I get this Starting stunnel: file /etc/stunnel/stunnel.conf line 9: Specified option name is not valid here If I comment out the xforwardedfor parameter, it starts. Any help is greatly appreciated. Thank you, Jill Rochelle

2 2

2 secure sockets TCP (Server / Client) as posted earlier last month by khaled
by Dick van Fulpen 04 Feb '08

04 Feb '08

> Hi, > > I'm having the same question (and problem) as posted by khaled on Mon > Jan 7 19:31:03 CET 2008, subject [stunnel-users] 2 secure sockets TCP > (Server / Client). > > I think I have to send data to stunnel on the client side by a write > to localhost:1974 where stunnel is configured with this stunnel.conf: > ; Sample stunnel configuration file by Michal Trojnara 2002-2006 > ; Some options used here may not be adequate for your particular > configuration > ; Certificate/key is needed in server mode and optional in client mode > ; The default certificate is provided only for testing and should not > ; be used in a production environment > cert = stunnel.pem > ;key = stunnel.pem > > ; Some performance tunings > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > > ; Use it for client mode > ;client = yes > > ; Service-level configuration > [sip2s] > accept = 1975 > connect = 127.0.0.1:1974 > > To start stunnel only works with just stunnel. > Started from batch or in dos-box with the options -d -r etc. as > explained in the documentation does not work, error message: file not > found) > BTW: I'm using Windows XP Professional on both client and server. > Client IP = 10.4.208.159, server IP = 10.4.208.102 for a test set-up. > > On the server side it gives the same error when started with options. > stunnel.conf on the server is this: > ; Sample stunnel configuration file by Michal Trojnara 2002-2006 > ; Some options used here may not be adequate for your particular > configuration > ; Certificate/key is needed in server mode and optional in client mode > ; The default certificate is provided only for testing and should not > be used in a production environment > cert = stunnel.pem > ;key = stunnel.pem > > ; Some performance tunings > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > > ; Some debugging stuff useful for troubleshooting > debug = 7 > output = stunnel.log > > ; Use it for client mode > client = yes > > ; Service-level configuration > [sip2s] > accept = 1975 > connect = 127.0.0.1:1974 > > Reading data on TCP/IP 1974 does not display anything when the client > is sending data to localhost:1974 which is encrypted by stunnel an > forwarded on TCP/IP 1975. > At least that is my intention, but something I do is wrong. > Data sending unencrypted between client an server works fine, so > connection is possible. > > Any idea what I should do to let it work like this: > - a client sends data to a server and reads the response from this > server. > - all data on the network between client an server should be > encrypted. > As far as I could find in the stunnel manual it should be able to use > stunnel for this. > In the manual is an example using foo (client) and bar (server). > > Dick van Fulpen | Senior Software Engineer | Infor | Tel:+31 (0)73 > 6205 222 | Dick.vanFulpen(a)infor.com | Infor | Postbus 2067 | 5202 CB > 's-Hertogenbosch | Nederland | Bezoek onze website www.vubis-smart.com > | >

1 0

Connection Refused locally, SSL Proxy.
by Avner Peled 01 Feb '08

01 Feb '08

Hello! I have a game client connecting to a server using TCP over SSL. I'm trying to setup an SSL proxy using 2 stunnels on the same server like so: Client --> Stunnel-Server:20001 (decrypt)--> Stunnel-Client:901 (encrypt)---> Server:20001 Using latest stunnel-4.21-r1 on Gentoo kernel 2.6.22-gentoo-r9 I have a weird problem. The first connection after starting both stunnels works fine. Afterwards every time the client tries to connect, it gets refused once. Then it tries again and gets connected. Then the next time it will once again get refused once and then connect. I have looked at the logs and it seems the Stunnel server (the first) gets Connection Refused from localhost:901. Then it tries again and gets connected. The Stunnel client (the second) doesn't even mention the first failing local connection in the log. I have provided the config files of both stunnels and a log with debug = 7 (server ip's are censored) I have also tried the following config tweaks with no luck socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 TIMEOUTclose = 0 retry = yes Any help would be appreciated. Thanks!! -- Avner Peled. avnerus(a)gmail.com

1 1

debug level 7
by Damian Wiest 25 Jan '08

25 Jan '08

Hi, I'm new to stunnel and have run into some unexpected behavior. I'm inclined to think it's a bug, but perhaps I'm misunderstanding some things. I'm running FreeBSD 6.2 with stunnel 4.21 (installed via ports collection). I'm able to configure things properly when using stunnel in daemon mode, but am unable to run stunnel from inetd when I set the debug level to 7. Here's the config file I'm using, cert = /usr/local/CA/stunnel.pem key = /usr/local/CA/stunnel.pem sslVersion = all pid = /var/log/swat-stunnel.pid debug = 7 output = /tmp/stunnel.log exec = /usr/local/sbin/swat execargs = swat ...and the inetd.conf entry, swat stream tcp nowait.400 root /usr/local/sbin/stunnel stunnel /usr/local/etc/stunnel/swat.conf If I connect to port 901, some debug messages are written to the socket, dwiest(a)dfwdamian.vail: telnet localhost 901 Trying ::1... telnet: connect to address ::1: Connection refused Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. 2008.01.22 22:19:09 LOG7[81278:134651904]: Snagged 64 random bytes from /home/dwiest/.rnd 2008.01.22 22:19:09 LOG7[81278:134651904]: Wrote 1024 new random bytes to /home/dwiest/.rnd 2008.01.22 22:19:09 LOG7[81278:134651904]: RAND_status claims sufficient entropy for the PRNG 2008.01.22 22:19:09 LOG7[81278:134651904]: PRNG seeded successfully 2008.01.22 22:19:09 LOG7[81278:134651904]: Certificate: /usr/local/CA/stunnel.pem 2008.01.22 22:19:09 LOG7[81278:134651904]: Certificate loaded 2008.01.22 22:19:09 LOG7[81278:134651904]: Key file: /usr/local/CA/stunnel.pem 2008.01.22 22:19:09 LOG7[81278:134651904]: Private key loaded 2008.01.22 22:19:09 LOG7[81278:134651904]: SSL context initialized for service stunnel ^] telnet> q Connection closed. ...and some are written to the file I specified, dwiest(a)dfwdamian.vail: cat /tmp/stunnel.log 2008.01.22 22:19:09 LOG5[81278:134651904]: stunnel 4.21 on i386-unknown-freebsd6.2 with OpenSSL 0.9.7e-p1 25 Oct 2004 2008.01.22 22:19:09 LOG5[81278:134651904]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP 2008.01.22 22:19:09 LOG7[81278:134651904]: stunnel started 2008.01.22 22:19:09 LOG7[81278:134651904]: FD 0 in non-blocking mode 2008.01.22 22:19:09 LOG7[81278:134651904]: FD 1 in non-blocking mode 2008.01.22 22:19:09 LOG7[81278:134651904]: stunnel permitted by libwrap from 127.0.0.1:53961 2008.01.22 22:19:09 LOG5[81278:134651904]: stunnel accepted connection from 127.0.0.1:53961 2008.01.22 22:19:09 LOG7[81278:134651904]: SSL state (accept): before/accept initialization 2008.01.22 22:19:11 LOG3[81278:134651904]: SSL_accept: Peer suddenly disconnected 2008.01.22 22:19:11 LOG5[81278:134651904]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.01.22 22:19:11 LOG7[81278:134651904]: stunnel finished (0 left) If I run stunnel in daemon mode with the debug level set to 7, then the messages are printed to stderr. Shouldn't they go to either syslog, or to the file specified with the "output" option? I dug through the source code and discovered that file_putline() is being called from parse_config() before open_log() has been called. This results in file_putline() writing to stderr instead of the specified output file. If this is known behavior, then it would be nice if the man page mentioned something. -Damian

3 3

rtmps with stunnel
by David Schulberg 24 Jan '08

24 Jan '08

Hi, A while ago I reported an issue with using rtmps on linux to connect to a Red5 server through stunnel. I am having no issues on Windows. The rtmps connection is made from Flash code running in the browser. The error reported on Linux was 'SSL_accept: Peer suddenly disconnected' which occurs straight after 'SSL state (accept): before/accept initialization' . Now I am having a similar problem trying to connect from a MAC. In this case I also get 'SSL_accept: Peer suddenly disconnected' which occurs straight after 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): before/accept initialization 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): SSLv3 read client hello A 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): SSLv3 write server hello A 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): SSLv3 write certificate A 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): SSLv3 write server done A 2008.01.24 16:39:31 LOG7[1608:2980]: SSL state (accept): SSLv3 flush data It seems to get further on the MAC but still fails. Regards, David

1 0

stunnel with Fedora core
by Andy 17 Jan '08

17 Jan '08

Hello everyone I am trying to configure stunnel on a Fedora Core machine, but have had little luck. The below configuration is in the /etc/stunne/stunnel.conf. I want this machine to send encrypted data to another machine via stunnel. My knowledge of stunnel is quite limited. chroot = /home/stunnel pid = /stunnel.pid setuid = stunnel setgid = stunne; debug = 7 output = /var/log/messages client = yes [ldap] accept = 389 connect = ldap.example.com:636 I have been told that this command will work. stunnel -c -d ldaps -r <LDAPSERVER>:ldaps This command does work under debian but does not work under fedora core. Can anyone help me with this? -- REGARDS, Andy

1 0

Debian Bug#460019: cert= produces output on stderr even with log= set
by Luis Rodrigo Gallardo Cruz 16 Jan '08

16 Jan '08

----- Forwarded message from Bryan Donlan <bd(a)fushizen.net> ----- Version: 3:4.21-1 With the following configuration: output = /srv/stun/log compression = zlib CApath = /srv/stun/keys cert = /srv/stun/server.pem connect = localhost:9999 verify = 3 The following is output on stderr: 2008.01.10 02:54:19 LOG5[11786:3083495088]: Peer certificate location /srv/stun/keys This confuses clients when stunnel is driven direcly from xinetd. This output, if it is produced at all, should be placed in the log file. ----- End forwarded message ----- The issue is that verify_init (verify.c:103) calls s_log(LOG_NOTICE, "Peer certificate location %s", section->ca_dir); apparently before the logfile is setup. Could the log file initialization be moved earlier in the starup sequence? Or maybe this log call's severity could be lowered.

2 6

Nessus crashes stunnel
by Joe Kemp 16 Jan '08

16 Jan '08

I am running some tests against my stunnel configuration with Nessus. I am able to get stunnel to exit silently when I run Nessus with the Nessus TCP Scan and Weak Supported SSL Cipher Suites test. I have played with a few different options but the process consistently ends when these tests are run together. Here is the config: Linux testssl.capwin.net 2.6.18-53.1.4.el5 #1 SMP Fri Nov 30 00:45:16 EST 2007 i686 i686 i386 GNU/Linux gcc version 4.1.2 20070626 (Red Hat 4.1.2-14) stunnel 4.20 on i686-pc-linux-gnu with OpenSSL 0.9.8b 04 May 2006 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv4 Auth:LIBWRAP OpenSSL 0.9.8b 04 May 2006 Tail of log file during nessus tests 2008.01.15 21:24:30 LOG5[10646:3086605200]: XMPP accepted connection from 10.102.11.250:41781 2008.01.15 21:24:30 LOG7[10646:3086605200]: SSL state (accept): before/accept initialization 2008.01.15 21:24:30 LOG7[10646:3086605200]: SSL alert (write): fatal: handshake failure 2008.01.15 21:24:30 LOG3[10646:3086605200]: SSL_accept: 1408A10B: error:1408A10B:SSL routines:SSL3_GET_CLIENT_HELLO:wrong version number 2008.01.15 21:24:30 LOG5[10646:3086605200]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2008.01.15 21:24:30 LOG7[10646:3086605200]: XMPP finished (0 left) 2008.01.15 21:24:30 LOG7[10646:3086608080]: XMPP accepted FD=7 from 10.102.11.250:41782 2008.01.15 21:24:30 LOG7[10646:3086605200]: XMPP started 2008.01.15 21:24:30 LOG7[10646:3086605200]: FD 7 in non-blocking mode 2008.01.15 21:24:30 LOG7[10646:3086605200]: TCP_NODELAY option set on local socket 2008.01.15 21:24:30 LOG7[10646:3086605200]: FD 8 in non-blocking mode 2008.01.15 21:24:30 LOG7[10646:3086605200]: FD 9 in non-blocking mode 2008.01.15 21:24:30 LOG7[10646:3086608080]: Cleaning up the signal pipe 2008.01.15 21:24:30 LOG6[10646:3086608080]: Child process 10676 finished with code 0 2008.01.15 21:24:30 LOG7[10646:3086605200]: Connection from 10.102.11.250:41782 permitted by libwrap 2008.01.15 21:24:30 LOG5[10646:3086605200]: XMPP accepted connection from 10.102.11.250:41782 2008.01.15 21:24:30 LOG7[10646:3086605200]: SSL state (accept): before/accept initialization Joe A. Kemp CapWIN Senior Systems Architect 6305 Ivy Lane Suite 300 Greenbelt, MD 20770 (P) 301-614-3727 (F) 301-614-0581

2 1

Re: [stunnel-users] stunnel automatically listening to extra ports: BAD
by Luis Rodrigo Gallardo Cruz 12 Jan '08

12 Jan '08

[You forgot to cc: the list] On Fri, Jan 11, 2008 at 03:24:32PM -0500, Alan Pinstein wrote: > On Jan 11, 2008, at 3:02 PM, Luis Rodrigo Gallardo Cruz wrote: >> On Fri, Jan 11, 2008 at 02:31:58PM -0500, Alan Pinstein wrote: >>> Hi All- >>> [ Stunnel listens on ports it shouldn't ] >>> >>> If I start up stunnel from the command line as "root" or another user >>> even, it only listens on the port listed in the conf file. >> >> Could you have some startup script that's automatically getting run >> with your logrotate scripts? >> > The problem happens the MOMENT stunnel starts. It has nothing to do with > the logrotate scripts. Oh, sorry, I got confused. (/me goes and re-reads the original mail ...) > I've done a bunch of debugging and can't figure out what's going on. > I have only one guess: stunnel automatically listens on any ports > that the process calling stunnel is listening on, in some sort of > attempt to seamlessly add SSL to existing daemons. I can't find any > docs or tell from the source code, but it's the only idea I can't > rule out... All this is in a *nix, right? The following is an educated guess: Open connections are open file descriptors, and fork()/exec() do not close open file descriptors. Thus, stunnel is inheriting the open connection. And, since it knows nothing about it, it does not close it or anything like that. So it's not that it activelly listens on it, but only that it does not bother to stop. You might check this theory by trying to connect to the port *after* apache is shut down. If I'm right you should get no answer at all. So, the solution is to have the file descriptors close when exec'ing stunnel. PHP or apache might have some option somewhere to do that (look for "close on exec" or something similarly named), but if not, you might have to write some sort of wrapper to do it. It's a messy thing, because AFAIK there's no clean way to do it short of for (i = 0; i < [some-hopefuly-large-enough-value]; i++) close(i); where the [large-enough-value] is determined by your system's file descriptor limit and your desire not to loop all the way to 2^32 or something. -- Rodrigo i-think-this-time-i-answered-the-question-asked Gallardo GPG-Fingerprint: 7C81 E60C 442E 8FBC D975 2F49 0199 8318 ADC9 BC28

5 4

Re: [stunnel-users] config socket SSL TCP
by khaled khaled 12 Jan '08

12 Jan '08

Hi,Can you send me please ( certificate+Key: Client & server) + CA.i want to try my server.The Client is a socket TCP developped on Openssl and C++.(I 'am sure that theclient work) but my config ???i want to generate a key for the client?the SSL client authenticate itself.Many thanks,FOR the server:-----BEGIN RSA PRIVATE KEY-----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-----END RSA PRIVATE KEY----------BEGIN CERTIFICATE-----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-----END CERTIFICATE-----For CA ROOT:-----BEGIN RSA PRIVATE KEY-----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-----END RSA PRIVATE KEY----------BEGIN CERTIFICATE-----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-----END CERTIFICATE-----FOR the Client-----BEGIN CERTIFICATE-----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-----END CERTIFICATE----------BEGIN RSA PRIVATE KEY-----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-----END RSA PRIVATE KEY----- _________________________________________________________________ Comme 9 millions de français, créez un compte Hotmail ! http://www.windowslive.fr/hotmail/default.asp

1 0

stunnel automatically listening to extra ports: BAD
by Alan Pinstein 11 Jan '08

11 Jan '08

Hi All- I am using stunnel to secure a connection between a local python script using telnetlib and a custom sockets-based server requiring SSL, from local port 4449 to remote port 4449. This is done via: stunnel /opt/www/domains/admin.showcasere.com/showcase/classes/ opensrs-php/stunnel-app.conf where the conf file is: > client = yes > pid = /opt/www/domains/admin.showcasere.com/runtime/stunnel.pid > debug = 7 > [telnet] > accept = 4449 > connect = admin.hostedemail.com:4449 Functionally, everything works for my application, but I am experiencing a bad side-effect. stunnel is ALSO setting up listeners on HTTP and HTTPS ports, and when my daily logrotate scripts run and HUP apache, stunnel steals the web server's ports and the server won't come back up! I had 7 hours of downtime today because of this. I've done a bunch of debugging and can't figure out what's going on. I have only one guess: stunnel automatically listens on any ports that the process calling stunnel is listening on, in some sort of attempt to seamlessly add SSL to existing daemons. I can't find any docs or tell from the source code, but it's the only idea I can't rule out... Here is the debug log of the startup of stunnel (which is run from an apache/php script): Jan 11 13:10:13 bigwoody stunnel: LOG5[13964:3086333632]: stunnel 4.14 on i386-redhat-linux-gnu PTHREAD+POLL+IPv6+LIBWRAP with OpenSSL 0.9.8a 11 Oct 2005 Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: RAND_status claims sufficient entropy for the PRNG Jan 11 13:10:13 bigwoody stunnel: LOG6[13964:3086333632]: PRNG seeded successfully Jan 11 13:10:13 bigwoody stunnel: LOG6[13964:3086333632]: file ulimit = 1024 (can be changed with 'ulimit -n') Jan 11 13:10:13 bigwoody stunnel: LOG6[13964:3086333632]: poll() used - no FD_SETSIZE limit for file descriptors Jan 11 13:10:13 bigwoody stunnel: LOG5[13964:3086333632]: 500 clients allowed Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: FD 31 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: FD 32 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: FD 33 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: SO_REUSEADDR option set on accept socket Jan 11 13:10:13 bigwoody stunnel: LOG7[13964:3086333632]: telnet bound to 0.0.0.0:4449 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086333632]: Created pid file /opt/www/domains/admin.showcasere.com/runtime/stunnel.pid Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086333632]: telnet accepted FD=34 from 127.0.0.1:48335 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: telnet started Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: FD 34 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: FD 35 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: FD 36 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086333632]: Cleaning up the signal pipe Jan 11 13:10:13 bigwoody stunnel: LOG6[13965:3086333632]: Child process 13967 finished with code 0 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: Connection from 127.0.0.1:48335 permitted by libwrap Jan 11 13:10:13 bigwoody stunnel: LOG5[13965:3086330784]: telnet connected from 127.0.0.1:48335 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: FD 35 in non-blocking mode Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: telnet connecting 216.40.42.6:4449 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: connect_wait: waiting 10 seconds Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: connect_wait: connected Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: Remote FD=35 initialized Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): before/connect initialization Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 write client hello A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 read server hello A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 read server certificate A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 read server done A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 write client key exchange A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 write change cipher spec A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 write finished A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 flush data Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL state (connect): SSLv3 read finished A Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 1 items in the session cache Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 1 client connects (SSL_connect()) Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 1 client connects that finished Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 client renegotiatations requested Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 server connects (SSL_accept()) Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 server connects that finished Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 server renegotiatiations requested Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 session cache hits Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 session cache misses Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: 0 session cache timeouts Jan 11 13:10:13 bigwoody stunnel: LOG6[13965:3086330784]: SSL connected: new session negotiated Jan 11 13:10:13 bigwoody stunnel: LOG6[13965:3086330784]: Negotiated ciphers: RC4-MD5 SSLv3 Kx=RSA Au=RSA Enc=RC4 (128) Mac=MD5 Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: SSL socket closed on SSL_read Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: Socket write shutdown Jan 11 13:10:13 bigwoody stunnel: LOG5[13965:3086330784]: Connection closed: 91 bytes sent to SSL, 73 bytes sent to socket Jan 11 13:10:13 bigwoody stunnel: LOG7[13965:3086330784]: telnet finished (0 left) And then, you can see what stunnel is listening on: [root@bigwoody custom_img]# lsof -i | grep stunnel stunnel 13965 apache 4u IPv4 156503478 TCP static-216.114.79.43.primarynetwork.com:http (LISTEN) stunnel 13965 apache 5u IPv4 156503480 TCP static-216.114.79.43.primarynetwork.com:https (LISTEN) stunnel 13965 apache 30u IPv4 156771437 TCP localhost.localdomain:51333->localhost.localdomain:9676 (ESTABLISHED) stunnel 13965 apache 33u IPv4 156846546 TCP *:privatewire (LISTEN) If I start up stunnel from the command line as "root" or another user even, it only listens on the port listed in the conf file. Does anyone have any idea what's going on here? How can I turn off this behavior? Thanks! Alan

2 1

Re: [stunnel-users] your mail
by Brian Hatch 10 Jan '08

10 Jan '08

On or about 2008-01-10 15:34 +0100, khaled45(a)free.fr directed: > Here attached files generated. > can you send me the configuration element of SSL and Stunnel or explain me how > to generate? You sent the private keys too - you shouldn't do this. You should create new keys when done debugging this problem. I saw only 2 keys - the CA and the stunnel.pem. Your configuration specifies 3 keys, the CA, stunnel.pem, and is attempting to verify the SSL client (browser? something else) against the CA via the 'verify=' line. What is the SSL client? Did you generate a key for it? Because your config is requiring the SSL client authenticate itself, and Stunnel is dropping the connection because that phase isn't working correctly. -- Brian Hatch "So, what's the problem?" Systems and "I don't know. Security Engineer *That's* the problem." http://www.ifokr.org/bri/ Every message PGP signed

1 0

Re: [stunnel-users] your mail
by Brian Hatch 09 Jan '08

09 Jan '08

At more or less 2008-01-09 18:09 +0100, khaled45(a)free.fr squalked: > ; It's often easier to use CAfile > CAfile = ca.pem > cert = userkey.pem > debug = 7 > client = no > [apache] > accept = 4433 > connect = 192.168.1.4:433 I'd suggest using the full path for your pem files above, just to be safe. Can you shoot the public keys of the client (which is not on this machine) and the ca.pem and the public key from userkey.pem (which is actually the server's key - you may wish to rename.) There are three pems involved: * the CA key (private) and cert (public) in ca.pem * the stunnel server's key (private) and cert (public) in userkey.pem. ('userkey' is a bad name here.) * the client's key (private) and cert (public) on 192.168.1.5. Can you send the public parts here so I can verify that they're signed by the appropritate folks? -- Brian Hatch "Everybody's cute! Systems and Everybody's cute! Security Engineer Even me. But in http://www.ifokr.org/bri/ purple I'm stunning!" Every message PGP signed

1 0

TCP (Server / Client)
by khaled khaled 09 Jan '08

09 Jan '08

Hi all, I am a new STUNNEL user,I will do 2 secure sockets TCP (Server / Client). I would like to know how to set 'stunnel.conf. I tried but it does not work. Description: ---------- I created a CA root and from this I have created 2 pairs (key and certificate) one for the server and one for the client. You can send me please a config file of the server. Many thanks, Stunnel.config (serveur) ----------------------- cert = server.crt key = serverkey.key CAfile = ca.crt ;[Server] accept = 192.168.1.4:4433 connect = 192.168.1.4:4433 _________________________________________________________________ Votre contact a choisi Hotmail, l'e-mail nouvelle génération. Créez un compte. http://www.windowslive.fr/hotmail/default.asp

3 6

2 secure sockets TCP (Server / Client).
by khaled khaled 07 Jan '08

07 Jan '08

Hi all,I am a new STUNNEL user,I will do 2 secure sockets TCP (Server / Client). I would like to know how to set 'stunnel.conf. I tried but it does not work. Description: ---------- I created a CA root and from this I have created 2 pairs (key and certificate) one for the server and one for the client. You can send me please a config file from a server? Many thanks, Stunnel.config (serveur) ----------------------- ; Sample stunnel configuration file by Michal Trojnara 2002-2006; Some options used here may not be adequate for your particular configuration ; Certificate/key is needed in server mode and optional in client mode; The default certificate is provided only for testing and should not; be used in a production environment;cert = stunnel.pem;key = stunnel.pemcert = server.crt key = serverkey.key ; Some performance tuningssocket = l:TCP_NODELAY=1socket = r:TCP_NODELAY=1 CAfile = ca.crt ;[Server]accept = 192.168.1.4:4433connect = 192.168.1.4:4433 KHALED Khaled Ingénieur de télécommunications et informatique _________________________________________________________________ Votre contact a choisi Hotmail, l'e-mail nouvelle génération. Créez un compte. http://www.windowslive.fr/hotmail/default.asp

1 0

Missing Something Basic on Config
by Steve 29 Dec '07

29 Dec '07

Using stunnel for NNTPS. Stunnel starts without any apparent problems (log looks normal), but simply doesn't intercept packets from my news reader. There is no activity in the log and sniffing the packets shows everything happening on 119. My understanding of the concept is that there are no configuration changes needed in the newsreader; IOW, I leave it pointing to the unsecured NNTP server with the port set to 119. I fear I'm missing something very obvious. BTW, I've confirmed the ability to use NNTPS with this particular server by using Outlook Express, which is compatible with NNTPS. Version: 4.20; I used the Windows binary. Running XP. Config: ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ; Some debugging stuff useful for troubleshooting debug = 7 output = stunnel.log ; Use it for client mode client = yes ; Service-level configuration [nntps] accept = 192.168.1.101:119 connect = secure.news.easynews.com:563 Log: 2007.12.28 22:24:46 LOG5[6640:6644]: stunnel 4.20 on x86-pc-mingw32-gnu with OpenSSL 0.9.8d 28 Sep 2006 2007.12.28 22:24:46 LOG5[6640:6644]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2007.12.28 22:24:46 LOG5[6640:5936]: No limit detected for the number of clients 2007.12.28 22:24:46 LOG7[6640:5936]: FD 196 in non-blocking mode 2007.12.28 22:24:46 LOG7[6640:5936]: SO_REUSEADDR option set on accept socket 2007.12.28 22:24:46 LOG7[6640:5936]: nntps bound to 192.168.1.101:119

1 0

Avoiding outlook express warning...
by fg 28 Dec '07

28 Dec '07

Hello all I use stunnel on a server with a stunnel.pem file generated with openSSL,( this file is not certified with verisign or others...) Each time i launch outlook express on a client a have a warning saying that the server is using a certificate who can't be verified.... I have to close this warning and after all is ok... Question Is it possible to skip this warning? perhaps with a public key approuved in outlook express??? thanks for help --------------------------------- Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail

3 2

stunnel
by peter griep 28 Dec '07

28 Dec '07

hello i'm using stunnel with download2day,but it's not working wih grabit can you tell what i must do to work stunnel with grabit. peter

3 2

1 tunnel client to few different servers
by Shalvi Ziv 26 Dec '07

26 Dec '07

All, As my PROXY server gets requests from end-users to be proxied to several different content providers (SMS servers), I was wondering how can encrypt the traffic between the PROXY to the SMS using stunnel, but still handle the differentiation to the various content providers. Meaning, my PROXY listens on port 9090 for all incoming SMS requests and each request should be addressed to different SMS server (i.e. different content provider) based on the number the message was sent to. See following example: Client = yes [SMPP] accept = 9090 (should always be this port) connect = <SMS IP>:442 Now, how can I do it with stunnel configuration (having the <SMS IP> flexibly assigned according to other parameters in the packet)? Thanks! zivs

2 1

Can stunnel be configured to listen for multiple IP addresses and access different email hosts?
by David Goldstein 21 Dec '07

21 Dec '07

2 1

Using stunnel as a non-root user
by T. S. Ferreira 20 Dec '07

20 Dec '07

I have a small web application running as a regular (non root) user on my port 6666. I would like to make it accessible through SSL on port 7777. In order to test it I used (as the same non root user) the following command: stunnel -D 7 -p mypemfile.pem -d 7777 -r localhost:6666 but it does not work. When I check with "ps aux | fgrep stunnel", no such program is running. However if I run it as root, it does work. I did not find in the documentation any restriction with this regard. Since my final application will run on a machine to whose root I do not have access, I would like to find a way to solve this problem. Any hint will be appreciated. -- tsf ----------------------------------------- My system data are: (1) Linux/Ubuntu 7.10, kernel 2.6.22-14-generic, (2) Output of "/usr/bin/stunnel4 -version": stunnel 4.20 on i486-pc-linux-gnu with OpenSSL 0.9.8e 23 Feb 2007 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /var/run/stunnel4.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none (3) Output due to '-D 7' option: 2007.12.20 10:00:45 LOG7[14820:3082589872]: Snagged 64 random bytes from /home/tomasz/.rnd 2007.12.20 10:00:45 LOG7[14820:3082589872]: Wrote 1024 new random bytes to /home/tomasz/.rnd 2007.12.20 10:00:45 LOG7[14820:3082589872]: RAND_status claims sufficient entropy for the PRNG 2007.12.20 10:00:45 LOG7[14820:3082589872]: PRNG seeded successfully 2007.12.20 10:00:45 LOG7[14820:3082589872]: Certificate: jai_ssl.pem 2007.12.20 10:00:45 LOG7[14820:3082589872]: Certificate loaded 2007.12.20 10:00:45 LOG7[14820:3082589872]: Key file: jai_ssl.pem 2007.12.20 10:00:45 LOG7[14820:3082589872]: Private key loaded 2007.12.20 10:00:45 LOG7[14820:3082589872]: SSL context initialized for service stunnel3 (4)"/usr/bin/stunnel" is a Perl script whose comments say " ... wrapper to use stunnel 3.x syntax in stunnel >=4.05 ..."; it is version 2.00 (2005.04.2).

1 1

Stunnel never really connects
by Pat Riehecky 20 Dec '07

20 Dec '07

I am trying to send email over stunnel from a client that doesn't support encryption to a server that does. The local socket opens just fine, but it never really seems to connect to the remote server. A tcpdump shows only three packets exchanged between the two systems (Syn, Syn/Ack, Ack). The behavior is the same regardless of which tls enabled port I point stunnel at. I am able to connect to these ports via telnet and send email so I know the connection can be negotiated. I am using the 4.21 build for windows (XP SP2) found on the main webpage. Help! --- stunnel.conf --- ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ; Workaround for Eudora bug ; my email client isn't eudora, but given how ; it doesn't support standards like encrypted ; transport..... options = DONT_INSERT_EMPTY_FRAGMENTS ; Some debugging stuff useful for troubleshooting debug = 7 ; Use it for client mode client = yes [smtp] accept = 127.0.0.1:25 connect = mailserver.mydomain.tld:587 ;connect = mailserver.mydomain.tld:465 ;connect = mailserver.mydomain.tld:25 protocol = smtp

2 1

'PIN' engine control command
by Ludolf Holzheid 14 Dec '07

14 Dec '07

Dear list, I'm trying to use stunnel on Windows with an USB token (`CryptoMate' from ACS). However, I can't manage to make stunnel ask for the token PIN: o Loading the private key works if I specify the PIN in the configuration file (`engineCtrl = PIN:...'). o If there is no `PIN' engine control option in the configuration file, loading the private key fails. The log message are: error stack: 26096080 : error:26096080:engine routines:ENGINE_load_private_key:failed loading private key error stack: 800050A0 : error:800050A0:Vendor defined:PKCS11_login:PIN incorrect Do I have to add a special engine control option to the configuration file in order to get asked for the PIN or is it intended to `just work'? Ludolf -- --------------------------------------------------------------- Ludolf Holzheid Tel: +49 621 339960 Bihl+Wiedemann GmbH Fax: +49 621 3392239 Floßwörthstraße 41 e-mail: lholzheid(a)bihl-wiedemann.de D-68199 Mannheim, Germany ---------------------------------------------------------------

2 2

rtmps with stunnel on linux
by David Schulberg 14 Dec '07

14 Dec '07

Hi, Cannot get rtmps working with Firefox (IceWeasel) on Linux. I have rtmps working fine with Firefox and IE on Windows XP. The error I see in stunnel (see below) when I try rtmps through a Linux browser is "peer suddenly disconnected". Also curious that there are 2 attempts made. 2007.12.13 16:31:41 LOG7[2624:1192]: https accepted FD=212 from 124.190.241.7:59728 2007.12.13 16:31:41 LOG7[2624:1192]: Creating a new thread 2007.12.13 16:31:41 LOG7[2624:1192]: New thread created 2007.12.13 16:31:41 LOG7[2624:2576]: https started 2007.12.13 16:31:41 LOG7[2624:2576]: FD 212 in non-blocking mode 2007.12.13 16:31:41 LOG7[2624:2576]: TCP_NODELAY option set on local socket 2007.12.13 16:31:41 LOG5[2624:2576]: https accepted connection from 124.190.241.7:59728 2007.12.13 16:31:41 LOG7[2624:2576]: SSL state (accept): before/accept initialization 2007.12.13 16:31:41 LOG3[2624:2576]: SSL_accept: Peer suddenly disconnected 2007.12.13 16:31:41 LOG5[2624:2576]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.12.13 16:31:41 LOG7[2624:2576]: https finished (0 left) 2007.12.13 16:31:41 LOG7[2624:1192]: https accepted FD=236 from 124.190.241.7:59729 2007.12.13 16:31:41 LOG7[2624:1192]: Creating a new thread 2007.12.13 16:31:41 LOG7[2624:1192]: New thread created 2007.12.13 16:31:41 LOG7[2624:2572]: https started 2007.12.13 16:31:41 LOG7[2624:2572]: FD 236 in non-blocking mode 2007.12.13 16:31:41 LOG7[2624:2572]: TCP_NODELAY option set on local socket 2007.12.13 16:31:41 LOG5[2624:2572]: https accepted connection from 124.190.241.7:59729 2007.12.13 16:31:41 LOG7[2624:2572]: SSL state (accept): before/accept initialization 2007.12.13 16:31:41 LOG3[2624:2572]: SSL_accept: Peer suddenly disconnected 2007.12.13 16:31:41 LOG5[2624:2572]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.12.13 16:31:41 LOG7[2624:2572]: https finished (0 left) Regards, David Schulberg

1 0

Re: [stunnel-users] * lencoboy Stunnel client and Centos 5 Apache
by Luis Rodrigo Gallardo Cruz 12 Dec '07

12 Dec '07

[Again, please send your responses to the list.] On Wed, Dec 12, 2007 at 06:18:58PM +0000, Mario Neè wrote: > > On Wed, Dec 12, 2007 at 05:00:08PM +0000, Mario Neè wrote: > > > > From: rodrigo(a)nul-unu.com > > > > This takes what is sent to local port 1443, encrypts it, and sends it > > > > like that to the remote 192.168.x.x:443 This means that the remote > > > > Apache needs to be listening on port 443 with ssl enabled there. Is > > > > this what you wanted? > > > > > > yes it's what i want. > > > > > > > Ok. Then I'd guess the problem is that apache is not expecting SSL on > > port 443. Read up on how to enable it. I'd guess a good start is > > http://httpd.apache.org/docs/2.0/mod/mod_ssl.html > > > if i try to connect to 192.168.x.x:443 with firefox everything > works, only with stunnel i get that error Do you try to connect with http://192.168.x.x:443 or with https://192.168.x.x:443 (note the 's')? -- Rodrigo Gallardo GPG-Fingerprint: 7C81 E60C 442E 8FBC D975 2F49 0199 8318 ADC9 BC28

1 0

Re: [stunnel-users] * lencoboy Stunnel client and Centos 5 Apache
by Luis Rodrigo Gallardo Cruz 12 Dec '07

12 Dec '07

[Please, respond to the list, not to the private mail of whoever answers you.] On Wed, Dec 12, 2007 at 05:00:08PM +0000, Mario Neè wrote: > > From: rodrigo(a)nul-unu.com > > On Wed, Dec 12, 2007 at 09:36:35AM +0000, Mario Neè wrote: > > > i'm trying to connect a stunnel client to the welcome page of Apache > > > in Centos 5 (Fresh install with no changes to Apache conf). > > > > > > command to create a client (on Feisty Fawn Ubuntu): > > > sudo stunnel -c -d 1443 -r 192.168.x.x:443 > > > > This takes what is sent to local port 1443, encrypts it, and sends it > > like that to the remote 192.168.x.x:443 This means that the remote > > Apache needs to be listening on port 443 with ssl enabled there. Is > > this what you wanted? > > yes it's what i want. > Ok. Then I'd guess the problem is that apache is not expecting SSL on port 443. Read up on how to enable it. I'd guess a good start is http://httpd.apache.org/docs/2.0/mod/mod_ssl.html -- Rodrigo Gallardo GPG-Fingerprint: 7C81 E60C 442E 8FBC D975 2F49 0199 8318 ADC9 BC28

1 0

newbie config not running
by fuzzy_4711 12 Dec '07

12 Dec '07

Hello list. I am new to stunnel and do have some problems to get it running. The server is a WinXPSP2 behind a router. The router has port forwarding enabled for port 9999 to the stunnel server (4.21). On the same box a vnc server is running on port 5900 which I like to tunnel. Firewall is opened. The netstat -a -n shows the server listening on the desired port (9999). My server config looks like that: cert = stunnel.pem ;key = stunnel.pem ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ; Workaround for Eudora bug ;options = DONT_INSERT_EMPTY_FRAGMENTS ; Authentication stuff ;verify = 2 ; Don't forget to c_rehash CApath ;CApath = certs ; It's often easier to use CAfile ;CAfile = certs.pem ; Don't forget to c_rehash CRLpath ;CRLpath = crls ; Alternatively you can use CRLfile ;CRLfile = crls.pem ; Some debugging stuff useful for troubleshooting ;debug = 7 ;output = stunnel.log ; Use it for client mode ;client = yes [pop3s] accept = 995 connect = 110 [imaps] accept = 993 connect = 143 [ssmtp] accept = 465 connect = 25 [VNC Server] accept = 9999 connect = 5900 The client side is a linux box with config setting: client = yes [VNC Server] accept = 5900 connect = 9999 I turned on debugging and stunnel.log seems to be okay - no error messages. But I am not able to get an connection. What am I doing wrong? Stefan

5 8

* lencoboy Stunnel client and Centos 5 Apache
by Mario Neè 12 Dec '07

12 Dec '07

Hello, i'm trying to connect a stunnel client to the welcome page of Apache in Centos 5 (Fresh install with no changes to Apache conf). command to create a client (on Feisty Fawn Ubuntu): sudo stunnel -c -d 1443 -r 192.168.x.x:443 if i try to connect to http://localhost:1443 and a look in /var/log/syslog i see this error: SSL_connect: error:1408F06B:SSL routines:SSL3_GET_RECORD:bad decompression. _________________________________________________________________ Ti piace giocare con le lettere? Prova ABCLive! http://messengergiochi.it.msn.com/

2 1

server reads ecrypted data on connect ?
by Nikolaj G. 10 Dec '07

10 Dec '07

Hello stunnel'ers, First of all - im a newbie into openssl and stunnel. Im trying to stunnel between my own java socket server and client. The stunnel is established, but the server reads encrypted data on connect. Since I dont send anything, Im surprised to see encrypted data in the server log. Could someone please help. Ive managed to create a root certificate, a request and signing it, using: http://www.eclectica.ca/howto/ssl-cert-howto.php Thanks in advance. /nikolaj PS see configuration below: ----------------------------------------------- Packages (Fedora Core 6): openssl-devel-0.9.8b-15.fc6 openssl-0.9.8b-15.fc6 stunnel-4.15-2 ----------------------------------------------- The SERVER stunnel.conf: ; Certificate/key is needed in server mode CAfile = /root/ssl/cacert.pem cert = /root/ssl/key-cert.pem pid = /root/ssl/stunnel.pid ; Some debugging stuff useful for troubleshooting debug = 7 foreground=yes #setuid = stunnel #setgid = stunnel [mystunnel] accept = ratata.cirque.dk:998 connect = 127.0.0.1:8200 ----------------------------------------------- The CLIENT stunnel.conf: ; Certificate/key is needed in server mode CAfile = /root/ssl/eclectica/cacert.pem pid = /root/ssl/eclectica/stunnel.pid ; Some debugging stuff useful for troubleshooting debug = 7 foreground=yes ; enable client mode client = yes [mystunnel] accept = 127.0.0.1:998 connect = ratata.cirque.dk:8200 ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs

1 0

HTTPS hardcoded redirects
by Ezio Ostorero 10 Dec '07

10 Dec '07

All, I have a problem similar to the one described in http://mirt.net/pipermail/stunnel-users/2006-October/001324.html i.e. I ***wish*** to use the recording functionalities of JMeter while accessing an apache-SSL-secured Tomcat web server. JMeter does not allow recording on SSL so I have to ... un-cipher my HTTPS sessions, stunnel looks like the right choice. Being an stunnel newbie, I started here: http://www.stunnel.org/examples/https_client.html , this HOWTO looks quite close to my configuration. We have a plain-vanilla tomcat server behind an apache/SSL, and I want to access this web application from an HTTP-only browser. So, I configure stunnel as a "client", I run it on my PC, with the following configuration: [psuedo-https] accept = 8080 connect = <server>:443 TIMEOUTclose = 0 I read it as follows: stunnel talk cleartext HTTP on the local 8080 port and forwards in crypted HTTPS on the <server> port 443 I set the URL in my browser to http://localhost:8080/oi/ and this happens: 1) Ethereal sez that my PC and <server> start an SSLv3 conversation, good, we're on track 2) I have an HTTP analyzer plugged in my browser that shows me the content of the first GET that is a redirect to an SSO server (on the same <server>:443 port) for user authentication https://<server>/ssoserver/login?service=...... 3) My browser then issue a GET to https://<server>/ssoserver/login?service=...... and is GAME OVER, my beloved stunnel is cleanly bypassed by the hardcoded https://<server>/ string Any suggestion? Is it a dead end? Any dirty trick I could play with? Such as running multiple stunnel instances, setting <server> = localhost in my hosts file etc.? Thanks, Ezio -- Ezio Ostorero, Catania Seltz e limone col sale. Arriminatu, non annacatu

1 0

Re: [stunnel-users] Problem compiling win32 stunnel.exe v 4.21
by butterspoon＠gmail.com 10 Dec '07

10 Dec '07

On 09/12/2007, Luis Rodrigo Gallardo Cruz <rodrigo(a)nul-unu.com> wrote: > On Sun, Dec 09, 2007 at 11:30:01PM +0000, Hidden wrote: > > Okay, so I'm trying to build stunnel.exe (ver 4.21) under Debian Linux > > 2.6.22.1 as a prelude to applying some patches and I'm getting > > stuck. > > Just to make sure I understand: You are trying to build *for* windows > *in* Debian. Yup, just like the documentation (INSTALL.W32) says: 1) install mingw32 cross-compiler o a Unix/Linux machine (native compilation on a Windows machine is possible, but not supported) [...] 5) type "make stunnel.exe" > You seem to be mixing up (the names of) unix libs with windows ones. Well, something is. I'm just doing it by the book! > Anyways, would it not be easier to compile in windows? Anyone else here build stunnel.exe? What did you use? T

1 0

Problem compiling win32 stunnel.exe v 4.21
by Hidden 09 Dec '07

09 Dec '07

Okay, so I'm trying to build stunnel.exe (ver 4.21) under Debian Linux 2.6.22.1 as a prelude to applying some patches and I'm getting stuck. I've followed the instructions in the makefile – Install mingw32; compile and install openssl; unpack and make stunnel.exe. mingw makes a bunch of object files for a while, but then at the link stage it bombs with i586-mingw32msvc/bin/ld: cannot find -lzdll make was when trying to execute the command: i586-mingw32msvc-gcc -s -o stunnel.exe file.obj client.obj log.obj options.obj protocol.obj network.obj resolver.obj ssl.obj ctx.obj verify.obj sthreads.obj stunnel.obj gui.obj resources.obj -L/usr/src/openssl-0.9.8g -lzdll -leay32 -lssl32 -lws2_32 -lgdi32 -mwindows eay32 is nowhere to be found either. It and ssl32 are clearly related to openssl, but I can't see any suitable files in either the openssl build directory (/usr/src/openssl-0.9.8g as advertised). zdll could be something to do with zlib (which I have, ver 1.2.3) and eay32 and ssl32 are obviously related to openssl, but the only libraries under openssl are libssl.a and libcrypto.a (although optimistically trying to bolt in libssl in place of ssl32 doesn't work). ws2_32 and gdi32 are windows sockets and GDI, for what it's worth. I can't find the required libs on the internet anywhere, and maybe it's all down to me being a novice with mingw. I do notice the tantalizingly named files zlib1.dll, libeay32.dll, and libssl32.dll in the precompiled Windows binary distribution of stunnel, but surely I can't use them in any way? I'd love some help from anyone who knows this system better! Thanks, Tony

2 1

SSL session caching with multiple clients through stunnel to an https server
by Paul McGarry 06 Dec '07

06 Dec '07

Hi there, I have an application where apache/php processes make SOAP requests over https to a single remote server (which is IIS I think). I am trying to cut down the time each request takes. In it's current configuration I think some time will be wasted as the php client and IIS server handshake to set up a new SSL connection for each request, thereby adding (at least) twice the network latency between the two servers to each request. (As far as I can tell there is no way to make php SOAP use some kind of persistant connection or SSL cache when talking to the server) I am wondering if I pushed the soap requests as http through stunnel whether stunnel would be more efficient with SSL connections or whether it will just create a new tunnel with a new ssl session for each request it receives? Does anyone have any advice? Paul

1 0

MS Outlook verify = 2 using problem
by FFT` 05 Dec '07

05 Dec '07

Good Day! I'm using Stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8g 19 Oct 2007 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 with OpenSSL 0.9.8g 19 Oct 2007 on Fedora-like Linux server. stunnel.conf consists: _______________________________________________________________________________ cert = /usr/local/etc/stunnel/server.crt key = /usr/local/etc/stunnel/server.pem ; Protocol version (all, SSLv2, SSLv3, TLSv1) ;sslVersion = SSLv3 ; Some security enhancements for UNIX systems - comment them out on Win32 chroot = /usr/local/var/stunnel/ setuid = nobody setgid = nobody ; PID is created inside chroot jail pid = /stunnel.pid ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ;compression = rle ; Workaround for Eudora bug ;options = DONT_INSERT_EMPTY_FRAGMENTS ; Authentication stuff verify = 2 ; Don't forget to c_rehash CApath ; CApath is located inside chroot jail ;CApath =/usr/local/etc/stunnel/certs ; It's often easier to use CAfile CAfile = /usr/local/etc/stunnel/ca.pem ; Don't forget to c_rehash CRLpath ; CRLpath is located inside chroot jail ;CRLpath = /crls ; Alternatively you can use CRLfile ;CRLfile = /usr/local/etc/stunnel/crl.pem ; Some debugging stuff useful for troubleshooting debug = 7 output = stunnel.log ; Service-level configuration [pop3s] accept = 995 connect = 110 ________________________________________________________________________ PKI consists: ROOT_CA CA MAIL_CON_CA SERVER CLIENT server.crt - Server certificate in PEM Format; server.pem - Server key in PEM Format with nopassword ca.pem - ROOT_CA, CA, MAIL_CON_CA certificates in PEM format. When I am using MS Outlook mail Client , there are some PROBLEM with connections to SERVER:995 stunnel.log consists: ______________________________________________ 2007.12.05 14:57:39 LOG5[16668:1074107776]: stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8g 19 Oct 2007 2007.12.05 14:57:39 LOG5[16668:1074107776]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 2007.12.05 14:57:39 LOG6[16668:1074107776]: file ulimit = 1024 (can be changed with 'ulimit -n') 2007.12.05 14:57:39 LOG6[16668:1074107776]: poll() used - no FD_SETSIZE limit for file descriptors 2007.12.05 14:57:39 LOG5[16668:1074107776]: 500 clients allowed 2007.12.05 14:57:39 LOG7[16668:1074107776]: FD 6 in non-blocking mode 2007.12.05 14:57:39 LOG7[16668:1074107776]: FD 8 in non-blocking mode 2007.12.05 14:57:39 LOG7[16668:1074107776]: FD 9 in non-blocking mode 2007.12.05 14:57:39 LOG7[16668:1074107776]: SO_REUSEADDR option set on accept socket 2007.12.05 14:57:39 LOG7[16668:1074107776]: pop3s bound to 0.0.0.0:995 2007.12.05 14:57:39 LOG7[16669:1074107776]: Created pid file /stunnel.pid 2007.12.05 14:57:48 LOG7[16669:1074107776]: pop3s accepted FD=10 from 192.168.1.205:2129 2007.12.05 14:57:48 LOG7[16669:1074228016]: pop3s started 2007.12.05 14:57:48 LOG7[16669:1074228016]: FD 10 in non-blocking mode 2007.12.05 14:57:48 LOG7[16669:1074228016]: TCP_NODELAY option set on local socket 2007.12.05 14:57:48 LOG5[16669:1074228016]: pop3s accepted connection from 192.168.1.205:2129 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): before/accept initialization 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): SSLv3 read client hello A 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): SSLv3 write server hello A 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): SSLv3 write certificate A 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): SSLv3 write certificate request A 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL state (accept): SSLv3 flush data 2007.12.05 14:57:48 LOG7[16669:1074228016]: SSL alert (write): fatal: handshake failure 2007.12.05 14:57:48 LOG3[16669:1074228016]: SSL_accept: 140890C7: error:140890C7:SSL routines:SSL3_GET_CLIENT_CERTIFICATE:peer did not return a certificate 2007.12.05 14:57:48 LOG5[16669:1074228016]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.12.05 14:57:48 LOG7[16669:1074228016]: pop3s finished (0 left) ______________________________________________ I load all Server, CA , and client certificate in SYSTEM, but there is NO Connection. Please help Me with Stunnel and Outlook using.

1 0

(no subject)
by Harmon, Paul 02 Dec '07

02 Dec '07

Ok Guys, I hate to be a nuisance. But I encountered an error coming out o the Stunnel configuration in LINUX. First what does it mean when Stunnel is in INETD mode? Does this imply certain configuration restrictions and why? Thanks, Paul Harmon

2 1

Help using Stunnel-4.XX
by Harmon, Paul 30 Nov '07

30 Nov '07

Hi Folks, Well I know a little more now than before. The version of Stunnel I'm using is Stunnel 4.05 running under Fedora 7.0. I am trying to setup a test scenario at the moment where I am running a stream socket between a client and a server, both mine since this is a test scenario. I need thisese programs to connect over a secure socket and the exchange of data is a custom application. I've read through Barry O'donovan's article and that seems to give me the information on how to configure and use Stunnel on the server side. It is not clear to me yet, what I need to do to the client side. Is there anyone out there who has done this type of thing and could give me some help? Thanks in advance, Paul Harmon

2 1

Problem with stunnel 4.20 on Fedora 7
by Andrew M Dunn 30 Nov '07

30 Nov '07

[Oh wise list folks... I haven't seen any responses to this. Does anybody have any experience either with using stunnel on Fedora 7, or with the particular error? I cannot find any other hints or reasons as to why the resolution is failing... Andy] ----original message------ I can't get stunnel to run on a freshly-installed Fedora 7 system. It fails with "Error resolving '0.0.0.0': servname is not supported for ai_socktype" but the usual other causes of that error (ie. ports not defined in /etc/services, bad settings in /etc/nsswitch.conf) are not the culprit this time. COMMAND LINE: /usr/sbin/stunnel /home/amdunn/security/stunnel.conf COMMAND OUTPUT: 2007.08.29 11:44:12 LOG3[13614:3086243520]: Error resolving '0.0.0.0': servname is not supported for ai_socktype (EAI_SERVICE) file /home/amdunn/security/stunnel.conf line 10: Failed to resolve accepting address STUNNEL.CONF FILE: ; ; stunnel file for testing 443 access to local http server ; cert=/home/amdunn/security/xxxxxxx key=/home/amdunn/security/yyyyyyy foreground=yes ; ; [sweb] accept=443 ; https port connect=80 ; standard httpd listening port TIMEOUTclose=0 ; must set this or IE waits 60sec on page end client=no STUNNEL -VERSION OUTPUT: stunnel 4.20 on i686-redhat-linux-gnu with OpenSSL 0.9.8b 04 May 2006 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /var/run/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none

2 1

zlib compression not working
by webmaster＠escoltesiguies.cat 29 Nov '07

29 Nov '07

I have configured both windows client and linux server (samba) with zlib compression: compression = zlib But I have transferred a txt file of 100Mb in 40 minutes and the same file compressed with gzip (22Mb) in 14 minutes. So I thing compression is not working. Client log tells: Compression enabled using zlib method and server log: Compression enabled using zlib method How can I verify comression is working? maybe compression is working but it is poor?

1 0

About stunnel 4.21 daemon mode
by wei hu 28 Nov '07

28 Nov '07

Dear I had successfully installed stunnel 4.21 on my server, but after I edited the file stunnel.conf and startup the stunnel, I can't find the stunnel via the "ps -ef" command.Then I have found such error message in the log file: 2007.11.27 07:42:28 LOG5[28550:3085354688]: stunnel 4.21 on i686-pc-linux-gnu wi th OpenSSL 0.9.7f 22 Mar 2005 2007.11.27 07:42:28 LOG5[28550:3085354688]: Threading:PTHREAD SSL:ENGINE Sockets :POLL,IPv6 Auth:LIBWRAP 2007.11.27 07:42:28 LOG6[28550:3085354688]: file ulimit = 1024 (can be changed w ith 'ulimit -n') 2007.11.27 07:42:28 LOG6[28550:3085354688]: poll() used - no FD_SETSIZE limit fo r file descriptors 2007.11.27 07:42:28 LOG5[28550:3085354688]: 500 clients allowed 2007.11.27 07:42:28 LOG7[28550:3085354688]: FD 9 in non-blocking mode 2007.11.27 07:42:28 LOG7[28550:3085354688]: FD 10 in non-blocking mode 2007.11.27 07:42:28 LOG7[28550:3085354688]: FD 11 in non-blocking mode 2007.11.27 07:42:28 LOG7[28550:3085354688]: SO_REUSEADDR option set on accept so cket 2007.11.27 07:42:28 LOG7[28550:3085354688]: sproxy bound to 0.0.0.0:83 2007.11.27 07:42:28 LOG3[28556:3085354688]: Failed to get GID for group nobody Do somebody knows how to get GID for group nobody? My nobody group is sured to work correctly with Squid. The config file about group: ; Some security enhancements for UNIX systems - comment them out on Win32 chroot = /usr/local/var/lib/stunnel/ setuid = nobody setgid = nobody ; PID is created inside chroot jail pid = /stunnel.pid The command "stunnel -version" response: [root@vps stunnel]# /usr/local/bin/stunnel -version stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.7f 22 Mar 2005 Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP Global options debug = 5 pid = /usr/local/var/run/stunnel/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options cert = /usr/local/etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH key = /usr/local/etc/stunnel/stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none The command "/usr/local/bin/stunnel" response: [root@vps stunnel]# /usr/local/bin/stunnel 2007.11.27 07:27:27 LOG7[13697:3085354688]: Snagged 64 random bytes from /root/.rnd 2007.11.27 07:27:27 LOG7[13697:3085354688]: Wrote 1024 new random bytes to /root/.rnd 2007.11.27 07:27:27 LOG7[13697:3085354688]: RAND_status claims sufficient entropy for the PRNG 2007.11.27 07:27:27 LOG7[13697:3085354688]: PRNG seeded successfully 2007.11.27 07:27:27 LOG7[13697:3085354688]: Certificate: /usr/local/etc/stunnel/stunnel.pem 2007.11.27 07:27:27 LOG7[13697:3085354688]: Certificate loaded 2007.11.27 07:27:27 LOG7[13697:3085354688]: Key file: /usr/local/etc/stunnel/stunnel.pem 2007.11.27 07:27:27 LOG7[13697:3085354688]: Private key loaded 2007.11.27 07:27:27 LOG7[13697:3085354688]: SSL context initialized for service sproxy Best Regards, HuWei

2 2

Re: [stunnel-users] More questions on RDP and port forwarding
by richard.woodman＠cox.net 28 Nov '07

28 Nov '07

Carter, Thanks. With the exception of the client's loopback address in the hosts file, I have a very similar configuration. I have created an inbound rule (Custom service with an inbound rule declaration but no outbound rule) that looks just like my inbound RDP rules - I even reviewed the configuration file and the stunnel rule had the same commands as the RDP rule. However, RDP works through the firewall but stunnel does not. My Windows firewall has an exception for stunnel with a scope of "Internet" and the stunnel server works from within the corporate network. Meaning, I think, that the Windows firewall is not denying the connection (at least from within the same subnet). I am not a Windows AD guy, but is there a way to prevent network connections without using the Windows firewall? I mean, can a GPO be created that prevents connections from IP addresses not on the same subnet but that does not use the OS's firewall? If so, this could be the issue because from what I can see, neither the Watchguard nor the Windows firewalls are blocking the connection; yet I still cannot connect. Richard Carter Browne wrote: I use stunnel to protect RDP for a couple of sites using a mix of Watchguard Edge and V series firewalls. For computer ABC that I want to connect to, I create an entry in the hosts file: 127.0.0.n ABCs (where n is greater than 1) On the client side I have an entry: [ABC-RDP] accept = ADCs:12345 connect = ABC:54321 client = yes On the server sid I have an entry: [RDP-IN] accept = 54321 connect = 3389 client = no Port 54321 is enabled in both the Watchguard and the Windows firewalls. Using the 127.n.n.n ports are not processed by the firewalls. You can use 127.0.0.1 for everything, but I needed to connect to more that one host and wanted a standard setup. I have had a number of users confused by this setup whereby the program references a local port to connect to a remote computer. For stunnel, it is the connect string that determines the destination, so any local port works fine for the accept string. Carter

2 1

Problems accessing samba
by webmaster＠escoltesiguies.cat 27 Nov '07

27 Nov '07

I have a winXP machine which should connect to a linux samba. Everything installed following http://wiki.netbsd.se/index.php/How_to_secure_samba_with_stunnel but it don't connect. I have both loglevels set to 7, and as serverlog don't change I thing the error is at clientside. stunnel.conf is: client=yes debug = 7 [smb] accept=222.222.222.222:139 connect=85.48.253.144:8888 And the client log is: 2007.11.27 12:46:58 LOG7[3716:720]: RAND_status claims sufficient entropy for the PRNG 2007.11.27 12:46:58 LOG7[3716:720]: PRNG seeded successfully 2007.11.27 12:46:58 LOG7[3716:720]: Certificate: stunnel.pem 2007.11.27 12:46:58 LOG7[3716:720]: Certificate loaded 2007.11.27 12:46:58 LOG7[3716:720]: Key file: stunnel.pem 2007.11.27 12:46:58 LOG7[3716:720]: Private key loaded 2007.11.27 12:46:58 LOG7[3716:720]: SSL context initialized for service smb 2007.11.27 12:46:58 LOG5[3716:720]: stunnel 4.21 on x86-pc-mingw32-gnu with OpenSSL 0.9.8g 19 Oct 2007 2007.11.27 12:46:58 LOG5[3716:720]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2007.11.27 12:46:58 LOG5[3716:3292]: No limit detected for the number of clients 2007.11.27 12:46:59 LOG7[3716:3292]: FD 216 in non-blocking mode 2007.11.27 12:46:59 LOG7[3716:3292]: SO_REUSEADDR option set on accept socket 2007.11.27 12:46:59 LOG7[3716:3292]: smb bound to 222.222.222.222:139 2007.11.27 12:47:15 LOG7[3716:3292]: smb accepted FD=248 from 88.31.107.217:1625 2007.11.27 12:47:15 LOG7[3716:3292]: Creating a new thread 2007.11.27 12:47:15 LOG7[3716:3292]: New thread created 2007.11.27 12:47:15 LOG7[3716:4028]: smb started 2007.11.27 12:47:15 LOG7[3716:4028]: FD 248 in non-blocking mode 2007.11.27 12:47:15 LOG7[3716:4028]: TCP_NODELAY option set on local socket 2007.11.27 12:47:15 LOG5[3716:4028]: smb accepted connection from 88.31.107.217:1625 2007.11.27 12:47:15 LOG7[3716:4028]: FD 272 in non-blocking mode 2007.11.27 12:47:15 LOG7[3716:4028]: smb connecting 85.48.253.144:8888 2007.11.27 12:47:15 LOG7[3716:4028]: connect_wait: waiting 10 seconds 2007.11.27 12:47:25 LOG7[3716:3292]: smb accepted FD=308 from 88.31.107.217:1628 2007.11.27 12:47:25 LOG7[3716:3292]: Creating a new thread 2007.11.27 12:47:25 LOG7[3716:3292]: New thread created 2007.11.27 12:47:25 LOG7[3716:3012]: smb started 2007.11.27 12:47:25 LOG7[3716:3012]: FD 308 in non-blocking mode 2007.11.27 12:47:25 LOG7[3716:3012]: TCP_NODELAY option set on local socket 2007.11.27 12:47:25 LOG5[3716:3012]: smb accepted connection from 88.31.107.217:1628 2007.11.27 12:47:25 LOG7[3716:3012]: FD 340 in non-blocking mode 2007.11.27 12:47:25 LOG7[3716:3012]: smb connecting 85.48.253.144:8888 2007.11.27 12:47:25 LOG7[3716:3012]: connect_wait: waiting 10 seconds 2007.11.27 12:47:25 LOG6[3716:4028]: connect_wait: s_poll_wait timeout 2007.11.27 12:47:25 LOG5[3716:4028]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.11.27 12:47:25 LOG7[3716:4028]: smb finished (1 left) 2007.11.27 12:47:35 LOG6[3716:3012]: connect_wait: s_poll_wait timeout 2007.11.27 12:47:35 LOG5[3716:3012]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.11.27 12:47:35 LOG7[3716:3012]: smb finished (0 left) Where is the problem? Thanks

1 1

VeriSign Intermediate CA Certificate issue.
by Unai Rodriguez 27 Nov '07

27 Nov '07

Dear All, I am trying to set an https layer using stunnel. I am having problems with the intermediary certificate since it appears broken to the user. You may find a screenshot here: http://u-journal.org/stunnel/Root_Cert_Err.jpg My stunnel version is: 3.26 running on Debian 3.1 I run stunnel as standalone with this parameters: stunnel -d 10.123.16.103:443 -D 7 -R /etc/ssl/certs/stunnel.rnd \ -p /etc/ssl/certs/stunnel.pem -o /var/log/stunnel/stunnel.log \ -r 10.123.16.103:80 -N https You may find the output of my stunnel -V here: http://pastebin.ca/793495 uname -a: Linux lb2-dp-vm 2.6.8-3-686 #1 Tue Dec 5 21:26:38 UTC 2006 i686 GNU/Linux openssl version: OpenSSL 0.9.7e 25 Oct 2004 So, if I create my /etc/ssl/certs/stunnel.pem file like this: -----BEGIN RSA PRIVATE KEY----- rsa_priv_key -----END RSA PRIVATE KEY----- -----BEGIN CERTIFICATE----- secured.behringer.com_certificate -----END CERTIFICATE----- -- OR -- -----BEGIN RSA PRIVATE KEY----- rsa_priv_key -----END RSA PRIVATE KEY----- -----BEGIN CERTIFICATE----- secured.behringer.com_certificate -----END CERTIFICATE----- -----BEGIN CERTIFICATE----- VeriSign Intermediate CA Certificate (https://www.verisign.com.sg/support/ssl/install/intermediate/secure.shtml) -----END CERTIFICATE----- I get the mentioned error. If I create the /etc/ssl/certs/stunnel.pem like this: -----BEGIN RSA PRIVATE KEY----- rsa_priv_key -----END RSA PRIVATE KEY----- -----BEGIN CERTIFICATE----- VeriSign Intermediate CA Certificate (https://www.verisign.com.sg/support/ssl/install/intermediate/secure.shtml) -----END CERTIFICATE----- -----BEGIN CERTIFICATE----- secured.behringer.com_certificate -----END CERTIFICATE----- I get the error: "[...] X509_check_private_key:key values mismatch" You may see the log here: http://pastebin.ca/793500 Any clues on what am I doing wrong? Thank you so much! unai

1 1

Re: [stunnel-users] More questions on RDP and port forwarding
by richard.woodman＠cox.net 26 Nov '07

26 Nov '07

OK, I took my configs, my certs, etc. and put them on two boxes in the lab - then I ensured I could use stunnel to connect using VNC. I then put a NAT box (Linksys BEFSR81) between the two and configured the Linksys to provide port forwarding. I then reconnected and everything worked first time. Therefore, I can only assume that either port-forwarding is misconfigured on my Watchguard SOHO 6tc -or- that the Windows firewall is causing problems (neither lab box has a firewall running). I'll keep you posted on what I find. Thanks. Richard My original post: I did read through the archives but I cannot determine how to get Stunnel working through the firewall. Here is what I wish to do: 1. Tunnel Windows Remote Desktop through stunnel. 2. I wish to connect from home to work; I have access to the firewall at work. Here's what I've done: 1. Installed stunnel on Windows XP at home and at work. I have self-signed certificates and am using verify = 3 (on both computers). Cacert.pem has the CA cert, the work cert, and the home cert in a single file. The server-cert.pem has the work computer's key and cert while the client-1-cert.pem (home computer) has it's own key and cert. 2. Stunnel at home has client = yes, stunnel at work has this commented out. Stunnel at work will become a "server" where multiple clients connect via stunnel and that single computer makes multiple RDP connections. Client (home) computer has [rdp1] accept = 4391 connect = <work outside interface IP>:44391 Server (work) computer has [rdp2] accept = 44391 connect = <work computer name>:3392 If I try this at work from within the corporate network (change the client connect string to the stunnel server's IP or hostname), then everything works fine. However, once I try from outside the work network, nothing works. Firewall is a Watchguard SOHO 6tc and I have a inbound rule permitting 44391 and directing it to X.X.X.52 (the stunnel server). I also have other rules allowing RDP (on port 3392 for instance) directly to the computer I wish to control and those rules work. Essentially, RDP directly through the firewall works but stunnel through the firewall does not. I assume there is no traffic destined for .52 on 44391 because the log file on the server (with debug = 7) only shows the startup sequence and port binding (netstat -a shows I am listening on 44391). I also tried this at home on my Juniper 5XT and was unsuccessful. Please help. Richard

1 0

Re: [stunnel-users] More questions on RDP and port forwarding
by richard.woodman＠cox.net 26 Nov '07

26 Nov '07

Craig, Thanks. Are you advising that I put "client = yes" in the service declaration rather than the main configuration? I already have "client = yes" on the client side (in the main configuration) and that line commented out of the server's configuration (I assume the default is client = no). When I review the log on the client computer, I can see where it times out waiting on a response from the work firewall's outside IP address. I do not see any connection attempts on the server side and both stunnel instances (client and server) have debug = 7 and logging to stunnel.log. Richard > -----Original Message----- > From: stunnel-users-bounces(a)mirt.net [mailto:stunnel-users- > bounces(a)mirt.net] On Behalf Of Richard Woodman > Sent: 26 November 2007 07:44 AM > To: stunnel-users(a)mirt.net > Subject: [stunnel-users] More questions on RDP and port forwarding > > I did read through the archives but I cannot determine how to get > Stunnel > working through the firewall. Here is what I wish to do: > > 1. Tunnel Windows Remote Desktop through stunnel. > 2. I wish to connect from home to work; I have access to the firewall > at > work. > > Here's what I've done: > > 1. Installed stunnel on Windows XP at home and at work. I have self- > signed > certificates and am using verify = 3 (on both computers). Cacert.pem > has > the CA cert, the work cert, and the home cert in a single file. The > server-cert.pem has the work computer's key and cert while the > client-1-cert.pem (home computer) has it's own key and cert. > 2. Stunnel at home has client = yes, stunnel at work has this > commented > out. Stunnel at work will become a "server" where multiple clients > connect > via stunnel and that single computer makes multiple RDP connections. > > Client (home) computer has > Try changing the client config to the following: [rdp1] client = yes accept = 4391 connect = <work outside interface IP>:44391 > > Server (work) computer has > > [rdp2] > accept = 44391 > connect = <work computer name>:3392 > > If I try this at work from within the corporate network (change the > client > connect string to the stunnel server's IP or hostname), then everything > works fine. However, once I try from outside the work network, nothing > works. Firewall is a Watchguard SOHO 6tc and I have a inbound rule > permitting 44391 and directing it to X.X.X.52 (the stunnel server). I > also > have other rules allowing RDP (on port 3392 for instance) directly to > the > computer I wish to control and those rules work. Essentially, RDP > directly > through the firewall works but stunnel through the firewall does not. > I > assume there is no traffic destined for .52 on 44391 because the log > file on > the server (with debug = 7) only shows the startup sequence and port > binding > (netstat -a shows I am listening on 44391). I also tried this at home > on my > Juniper 5XT and was unsuccessful. Please help. > > Richard > > > _______________________________________________ > stunnel-users mailing list > stunnel-users(a)mirt.net > http://stunnel.mirt.net/mailman/listinfo/stunnel-users Cheers, Craig

1 0

More questions on RDP and port forwarding
by Richard Woodman 26 Nov '07

26 Nov '07

I did read through the archives but I cannot determine how to get Stunnel working through the firewall. Here is what I wish to do: 1. Tunnel Windows Remote Desktop through stunnel. 2. I wish to connect from home to work; I have access to the firewall at work. Here's what I've done: 1. Installed stunnel on Windows XP at home and at work. I have self-signed certificates and am using verify = 3 (on both computers). Cacert.pem has the CA cert, the work cert, and the home cert in a single file. The server-cert.pem has the work computer's key and cert while the client-1-cert.pem (home computer) has it's own key and cert. 2. Stunnel at home has client = yes, stunnel at work has this commented out. Stunnel at work will become a "server" where multiple clients connect via stunnel and that single computer makes multiple RDP connections. Client (home) computer has [rdp1] accept = 4391 connect = <work outside interface IP>:44391 Server (work) computer has [rdp2] accept = 44391 connect = <work computer name>:3392 If I try this at work from within the corporate network (change the client connect string to the stunnel server's IP or hostname), then everything works fine. However, once I try from outside the work network, nothing works. Firewall is a Watchguard SOHO 6tc and I have a inbound rule permitting 44391 and directing it to X.X.X.52 (the stunnel server). I also have other rules allowing RDP (on port 3392 for instance) directly to the computer I wish to control and those rules work. Essentially, RDP directly through the firewall works but stunnel through the firewall does not. I assume there is no traffic destined for .52 on 44391 because the log file on the server (with debug = 7) only shows the startup sequence and port binding (netstat -a shows I am listening on 44391). I also tried this at home on my Juniper 5XT and was unsuccessful. Please help. Richard

3 2

I'm very much a newbie to stunnel
by Harmon, Paul 26 Nov '07

26 Nov '07

Hi Guys, I'm very new to Stunnel. In fact I have no experience with it what so ever. My situation is this: I have server program and a client at the moment which know nothing about SSL. Yet the powers that be insist that my two machines, if they must use sockets, do so using SSL. After reading through the distribution, I found this very interesting program called Stunnel which sounds like it might solve my problem quite well. From what I've read so far, I gather this program can be used as an SSL wrapper for my server and my client. Would some kind soul please help me with the configuration process (the configuration file) so that my client and server can be safe and secure? My server uses a port which I have chosen (really quite arbitrarily). Basically I need the server to listen on that port and wait for incoming data which it will processs and in most cases not even reply. With normal sockets this is a no-brainer since Steven's book outlines it very well. If I understand the documentation I've read so far, Stunnel adds the encryption layer on top of that which I need. So how do I configure and invokestunnel in this environment? Sorry for stupid questions at the moment I just lack examples like this and confidence in using the program. Thanks, Paul

2 1

Certificate management
by Koenraad Lelong 23 Nov '07

23 Nov '07

Hi, Since a few months I'm playing with stunnel. Last month I installed a stunnel at a customer. This week he reported he could not connect any more. After testing here I discovered the CA cert had expired. I made this myself and I didn't see that it's only valid for one month. I made new certificates making sure they are valid for longer than one month, so this is solved for now. My question : are there good documents about management of certificates ? I'm specially interested in renewing the certificates. Regards, Koenraad Lelong.

2 1

some newbie questions
by Lluís Forns 23 Nov '07

23 Nov '07

Let me first explain what I want to do in order to see if stunnel is what I need: I have some laptops around the world and I want to let them login in my intranet. The login is versus a Samba server. So more or less the connection would be: Laptop (running winXP) --- router --- internet --- router --- linux ssh server(port 22 open) --- linux samba server Could stunnel help me achivieng this? any hint of how? I have read the documentation, but I am a little lost. thanks -- <http://www.escoltesiguies.cat>*Lluís Forns - Tècnic Informàtic * *Servei d'Informàtica* *Minyons Escoltes i Guies de Catalunya (MEG)* *Telèfon:* 93 590 27 00 *Fax:* 93 590 04 92 *www.escoltesiguies.cat* No m'imprimeixis si no és necessari. Protegim el medi ambient.

2 1

Re: [stunnel-users] Stunnel and Outlook Web Access Problems
by Craig Retief 20 Nov '07

20 Nov '07

I see that you have client = yes in the config file. When client = yes is set outside the service block, it becomes a global setting that tells all the services that it should tunnel that they are clients and not servers. If you specify the client = yes within a service eg. [pop3s] client = yes accept = 995 connect = 110 then stunnel knows that the specified service alone is a client. What I recommend is that you disable the client = yes in the global area and change you config file to something like this: I have removed the comments to shorten the config for explanation: socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 CRLpath = crls CRLfile = certnew.pem debug = 7 output = stunnel.log ;client = yes [https] accept = 443 connect = 192.168.0.5:443 TIMEOUTclose = 0 Try using this as your config and see if it works or not. Cheers Craig From: Dan Vespa [mailto:dan@sideshow.sytes.net] Sent: 19 November 2007 05:07 PM To: Craig Retief Subject: RE: [stunnel-users] Stunnel and Outlook Web Access Problems Craig, Here it is.... ; Sample stunnel configuration file by Michal Trojnara 2002-2006 ; Some options used here may not be adequate for your particular configuration ; Certificate/key is needed in server mode and optional in client mode ; The default certificate is provided only for testing and should not ; be used in a production environment ;cert = stunnel.pem ;key = stunnel.pem ; Some performance tunings socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 ; Workaround for Eudora bug ;options = DONT_INSERT_EMPTY_FRAGMENTS ; Authentication stuff ;verify = 2 ; Don't forget to c_rehash CApath ;CApath = crls ; It's often easier to use CAfile ;CAfile = certnew.pem ; Don't forget to c_rehash CRLpath CRLpath = crls ; Alternatively you can use CRLfile CRLfile = certnew.pem ; Some debugging stuff useful for troubleshooting debug = 7 output = stunnel.log ; Use it for client mode client = yes ; Service-level configuration ;[pop3s] ;accept = 995 ;connect = 110 ;[imaps] ;accept = 993 ;connect = 143 ;[ssmtp] ;accept = 465 ;connect = 25 [https] accept = 443 connect = 192.168.0.5:443 TIMEOUTclose = 0 ; vim:ft=dosini _____ From: Craig Retief Sent: Mon 19/11/2007 9:54 AM To: 'Dan Vespa' Cc: stunnel-users(a)mirt.net Subject: RE: [stunnel-users] Stunnel and Outlook Web Access Problems Please Post your stunnel configs. Makes life easier. Cheers Craig From: Dan Vespa [mailto:dan@sideshow.sytes.net] Sent: 19 November 2007 03:35 PM To: Craig Subject: RE: [stunnel-users] Stunnel and Outlook Web Access Problems Craig, I set the debug level higher and this is what my log outputs.... 2007.11.19 08:25:54 LOG7[1596:2656]: RAND_status claims sufficient entropy for the PRNG 2007.11.19 08:25:54 LOG7[1596:2656]: PRNG seeded successfully 2007.11.19 08:25:54 LOG7[1596:2656]: SSL context initialized for service https 2007.11.19 08:25:54 LOG5[1596:2656]: stunnel 4.21 on x86-pc-mingw32-gnu with OpenSSL 0.9.8g 19 Oct 2007 2007.11.19 08:25:54 LOG5[1596:2656]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2007.11.19 08:25:54 LOG5[1596:3012]: No limit detected for the number of clients 2007.11.19 08:25:54 LOG7[1596:3012]: FD 188 in non-blocking mode 2007.11.19 08:25:54 LOG7[1596:3012]: SO_REUSEADDR option set on accept socket 2007.11.19 08:25:54 LOG7[1596:3012]: https bound to 0.0.0.0:443 2007.11.19 08:26:32 LOG7[1596:3012]: https accepted FD=236 from 127.0.0.1:1409 2007.11.19 08:26:32 LOG7[1596:3012]: Creating a new thread 2007.11.19 08:26:32 LOG7[1596:3012]: New thread created 2007.11.19 08:26:32 LOG7[1596:2576]: https started 2007.11.19 08:26:32 LOG7[1596:2576]: FD 236 in non-blocking mode 2007.11.19 08:26:32 LOG7[1596:2576]: TCP_NODELAY option set on local socket 2007.11.19 08:26:32 LOG5[1596:2576]: https accepted connection from 127.0.0.1:1409 2007.11.19 08:26:32 LOG7[1596:2576]: FD 260 in non-blocking mode 2007.11.19 08:26:32 LOG7[1596:2576]: https connecting 192.168.0.5:443 2007.11.19 08:26:32 LOG7[1596:2576]: connect_wait: waiting 10 seconds 2007.11.19 08:26:32 LOG7[1596:2576]: connect_wait: connected 2007.11.19 08:26:32 LOG5[1596:2576]: https connected remote server from 192.168.0.24:1410 2007.11.19 08:26:32 LOG7[1596:2576]: Remote FD=260 initialized 2007.11.19 08:26:32 LOG7[1596:2576]: TCP_NODELAY option set on remote socket 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): before/connect initialization 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write client hello A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server hello A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server certificate A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server done A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write client key exchange A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write change cipher spec A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write finished A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 flush data 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read finished A 2007.11.19 08:26:32 LOG7[1596:2576]: 1 items in the session cache 2007.11.19 08:26:32 LOG7[1596:2576]: 1 client connects (SSL_connect()) 2007.11.19 08:26:32 LOG7[1596:2576]: 1 client connects that finished 2007.11.19 08:26:32 LOG7[1596:2576]: 0 client renegotiations requested 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server connects (SSL_accept()) 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server connects that finished 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server renegotiations requested 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache hits 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache misses 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache timeouts 2007.11.19 08:26:32 LOG6[1596:2576]: SSL connected: new session negotiated 2007.11.19 08:26:32 LOG6[1596:2576]: Negotiated ciphers: RC4-MD5 SSLv3 Kx=RSA Au=RSA Enc=RC4(128) Mac=MD5 2007.11.19 08:26:53 LOG7[1596:3012]: https accepted FD=288 from 127.0.0.1:1411 2007.11.19 08:26:53 LOG7[1596:3012]: Creating a new thread 2007.11.19 08:26:53 LOG7[1596:3012]: New thread created 2007.11.19 08:26:53 LOG7[1596:1060]: https started 2007.11.19 08:26:53 LOG7[1596:1060]: FD 288 in non-blocking mode 2007.11.19 08:26:53 LOG7[1596:1060]: TCP_NODELAY option set on local socket 2007.11.19 08:26:53 LOG5[1596:1060]: https accepted connection from 127.0.0.1:1411 2007.11.19 08:26:53 LOG7[1596:1060]: FD 312 in non-blocking mode 2007.11.19 08:26:53 LOG7[1596:1060]: https connecting 192.168.0.5:443 2007.11.19 08:26:53 LOG7[1596:1060]: connect_wait: waiting 10 seconds 2007.11.19 08:26:53 LOG7[1596:1060]: connect_wait: connected 2007.11.19 08:26:53 LOG5[1596:1060]: https connected remote server from 192.168.0.24:1412 2007.11.19 08:26:53 LOG7[1596:1060]: Remote FD=312 initialized 2007.11.19 08:26:53 LOG7[1596:1060]: TCP_NODELAY option set on remote socket 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): before/connect initialization 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write client hello A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 read server hello A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 read finished A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write change cipher spec A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write finished A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 flush data 2007.11.19 08:26:53 LOG7[1596:1060]: 1 items in the session cache 2007.11.19 08:26:53 LOG7[1596:1060]: 2 client connects (SSL_connect()) 2007.11.19 08:26:53 LOG7[1596:1060]: 2 client connects that finished 2007.11.19 08:26:53 LOG7[1596:1060]: 0 client renegotiations requested 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server connects (SSL_accept()) 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server connects that finished 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server renegotiations requested 2007.11.19 08:26:53 LOG7[1596:1060]: 1 session cache hits 2007.11.19 08:26:53 LOG7[1596:1060]: 0 session cache misses 2007.11.19 08:26:53 LOG7[1596:1060]: 0 session cache timeouts 2007.11.19 08:26:53 LOG6[1596:1060]: SSL connected: previous session reused 2007.11.19 08:26:53 LOG7[1596:1060]: SSL socket closed on SSL_read 2007.11.19 08:26:53 LOG7[1596:1060]: Socket write shutdown 2007.11.19 08:26:53 LOG5[1596:1060]: Connection closed: 70 bytes sent to SSL, 164 bytes sent to socket 2007.11.19 08:26:53 LOG7[1596:1060]: https finished (1 left) 2007.11.19 08:27:53 LOG3[1596:2576]: readsocket: Connection reset by peer (WSAECONNRESET) (10054) 2007.11.19 08:27:53 LOG5[1596:2576]: Connection reset: 1102 bytes sent to SSL, 1241 bytes sent to socket 2007.11.19 08:27:53 LOG7[1596:2576]: https finished (0 left) Does this make any sense to you???? Dan _____ From: Craig Sent: Mon 19/11/2007 1:34 AM To: stunnel-users(a)mirt.net Subject: Re: [stunnel-users] Stunnel and Outlook Web Access Problems OWA rewrites the URL when you log in and I have found that it causes problems when doing that with a Stunnel connection. What I think might be happening is that your initial connection says http://server.tld/ then when you log in OWA rewrites the URL to https://server.tld/ What I think is happening is that your stunnel is listening on port 80 and when you log in, OWA changes the listening port in the browser to port 443. Try running the listening server (stunnel server) to listen on port 443 instead of port 80 and see if that fixes the problem. Cheers Craig From: stunnel-users-bounces(a)mirt.net [mailto:stunnel-users-bounces@mirt.net] On Behalf Of Dan Vespa Sent: 17 November 2007 07:21 PM To: stunnel-users(a)mirt.net Subject: [stunnel-users] Stunnel and Outlook Web Access Problems I am trying to use the latest version of stunnel to connect to OWA. I set stunnel to listen on port 80 (localhost) and then connect to myserverip on port 443. I get the login screen but can go no further after I Enter my credentials and Click OK. Checking the stunnel log screen it shows that a connect has been made? I don't understand why It won't connect any further?? OWA is on Exchange Server 2003 and I am using Windows XP SP2. I have also updated to Open SSL latest version. Thanks in Advance. Dan

1 0

Re: [stunnel-users] Stunnel and Outlook Web Access Problems
by Craig Retief 19 Nov '07

19 Nov '07

Please Post your stunnel configs. Makes life easier. Cheers Craig From: Dan Vespa [mailto:dan@sideshow.sytes.net] Sent: 19 November 2007 03:35 PM To: Craig Subject: RE: [stunnel-users] Stunnel and Outlook Web Access Problems Craig, I set the debug level higher and this is what my log outputs.... 2007.11.19 08:25:54 LOG7[1596:2656]: RAND_status claims sufficient entropy for the PRNG 2007.11.19 08:25:54 LOG7[1596:2656]: PRNG seeded successfully 2007.11.19 08:25:54 LOG7[1596:2656]: SSL context initialized for service https 2007.11.19 08:25:54 LOG5[1596:2656]: stunnel 4.21 on x86-pc-mingw32-gnu with OpenSSL 0.9.8g 19 Oct 2007 2007.11.19 08:25:54 LOG5[1596:2656]: Threading:WIN32 SSL:ENGINE Sockets:SELECT,IPv6 2007.11.19 08:25:54 LOG5[1596:3012]: No limit detected for the number of clients 2007.11.19 08:25:54 LOG7[1596:3012]: FD 188 in non-blocking mode 2007.11.19 08:25:54 LOG7[1596:3012]: SO_REUSEADDR option set on accept socket 2007.11.19 08:25:54 LOG7[1596:3012]: https bound to 0.0.0.0:443 2007.11.19 08:26:32 LOG7[1596:3012]: https accepted FD=236 from 127.0.0.1:1409 2007.11.19 08:26:32 LOG7[1596:3012]: Creating a new thread 2007.11.19 08:26:32 LOG7[1596:3012]: New thread created 2007.11.19 08:26:32 LOG7[1596:2576]: https started 2007.11.19 08:26:32 LOG7[1596:2576]: FD 236 in non-blocking mode 2007.11.19 08:26:32 LOG7[1596:2576]: TCP_NODELAY option set on local socket 2007.11.19 08:26:32 LOG5[1596:2576]: https accepted connection from 127.0.0.1:1409 2007.11.19 08:26:32 LOG7[1596:2576]: FD 260 in non-blocking mode 2007.11.19 08:26:32 LOG7[1596:2576]: https connecting 192.168.0.5:443 2007.11.19 08:26:32 LOG7[1596:2576]: connect_wait: waiting 10 seconds 2007.11.19 08:26:32 LOG7[1596:2576]: connect_wait: connected 2007.11.19 08:26:32 LOG5[1596:2576]: https connected remote server from 192.168.0.24:1410 2007.11.19 08:26:32 LOG7[1596:2576]: Remote FD=260 initialized 2007.11.19 08:26:32 LOG7[1596:2576]: TCP_NODELAY option set on remote socket 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): before/connect initialization 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write client hello A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server hello A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server certificate A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read server done A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write client key exchange A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write change cipher spec A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 write finished A 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 flush data 2007.11.19 08:26:32 LOG7[1596:2576]: SSL state (connect): SSLv3 read finished A 2007.11.19 08:26:32 LOG7[1596:2576]: 1 items in the session cache 2007.11.19 08:26:32 LOG7[1596:2576]: 1 client connects (SSL_connect()) 2007.11.19 08:26:32 LOG7[1596:2576]: 1 client connects that finished 2007.11.19 08:26:32 LOG7[1596:2576]: 0 client renegotiations requested 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server connects (SSL_accept()) 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server connects that finished 2007.11.19 08:26:32 LOG7[1596:2576]: 0 server renegotiations requested 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache hits 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache misses 2007.11.19 08:26:32 LOG7[1596:2576]: 0 session cache timeouts 2007.11.19 08:26:32 LOG6[1596:2576]: SSL connected: new session negotiated 2007.11.19 08:26:32 LOG6[1596:2576]: Negotiated ciphers: RC4-MD5 SSLv3 Kx=RSA Au=RSA Enc=RC4(128) Mac=MD5 2007.11.19 08:26:53 LOG7[1596:3012]: https accepted FD=288 from 127.0.0.1:1411 2007.11.19 08:26:53 LOG7[1596:3012]: Creating a new thread 2007.11.19 08:26:53 LOG7[1596:3012]: New thread created 2007.11.19 08:26:53 LOG7[1596:1060]: https started 2007.11.19 08:26:53 LOG7[1596:1060]: FD 288 in non-blocking mode 2007.11.19 08:26:53 LOG7[1596:1060]: TCP_NODELAY option set on local socket 2007.11.19 08:26:53 LOG5[1596:1060]: https accepted connection from 127.0.0.1:1411 2007.11.19 08:26:53 LOG7[1596:1060]: FD 312 in non-blocking mode 2007.11.19 08:26:53 LOG7[1596:1060]: https connecting 192.168.0.5:443 2007.11.19 08:26:53 LOG7[1596:1060]: connect_wait: waiting 10 seconds 2007.11.19 08:26:53 LOG7[1596:1060]: connect_wait: connected 2007.11.19 08:26:53 LOG5[1596:1060]: https connected remote server from 192.168.0.24:1412 2007.11.19 08:26:53 LOG7[1596:1060]: Remote FD=312 initialized 2007.11.19 08:26:53 LOG7[1596:1060]: TCP_NODELAY option set on remote socket 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): before/connect initialization 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write client hello A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 read server hello A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 read finished A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write change cipher spec A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 write finished A 2007.11.19 08:26:53 LOG7[1596:1060]: SSL state (connect): SSLv3 flush data 2007.11.19 08:26:53 LOG7[1596:1060]: 1 items in the session cache 2007.11.19 08:26:53 LOG7[1596:1060]: 2 client connects (SSL_connect()) 2007.11.19 08:26:53 LOG7[1596:1060]: 2 client connects that finished 2007.11.19 08:26:53 LOG7[1596:1060]: 0 client renegotiations requested 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server connects (SSL_accept()) 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server connects that finished 2007.11.19 08:26:53 LOG7[1596:1060]: 0 server renegotiations requested 2007.11.19 08:26:53 LOG7[1596:1060]: 1 session cache hits 2007.11.19 08:26:53 LOG7[1596:1060]: 0 session cache misses 2007.11.19 08:26:53 LOG7[1596:1060]: 0 session cache timeouts 2007.11.19 08:26:53 LOG6[1596:1060]: SSL connected: previous session reused 2007.11.19 08:26:53 LOG7[1596:1060]: SSL socket closed on SSL_read 2007.11.19 08:26:53 LOG7[1596:1060]: Socket write shutdown 2007.11.19 08:26:53 LOG5[1596:1060]: Connection closed: 70 bytes sent to SSL, 164 bytes sent to socket 2007.11.19 08:26:53 LOG7[1596:1060]: https finished (1 left) 2007.11.19 08:27:53 LOG3[1596:2576]: readsocket: Connection reset by peer (WSAECONNRESET) (10054) 2007.11.19 08:27:53 LOG5[1596:2576]: Connection reset: 1102 bytes sent to SSL, 1241 bytes sent to socket 2007.11.19 08:27:53 LOG7[1596:2576]: https finished (0 left) Does this make any sense to you???? Dan _____ From: Craig Sent: Mon 19/11/2007 1:34 AM To: stunnel-users(a)mirt.net Subject: Re: [stunnel-users] Stunnel and Outlook Web Access Problems OWA rewrites the URL when you log in and I have found that it causes problems when doing that with a Stunnel connection. What I think might be happening is that your initial connection says http://server.tld/ then when you log in OWA rewrites the URL to https://server.tld/ What I think is happening is that your stunnel is listening on port 80 and when you log in, OWA changes the listening port in the browser to port 443. Try running the listening server (stunnel server) to listen on port 443 instead of port 80 and see if that fixes the problem. Cheers Craig From: stunnel-users-bounces(a)mirt.net [mailto:stunnel-users-bounces@mirt.net] On Behalf Of Dan Vespa Sent: 17 November 2007 07:21 PM To: stunnel-users(a)mirt.net Subject: [stunnel-users] Stunnel and Outlook Web Access Problems I am trying to use the latest version of stunnel to connect to OWA. I set stunnel to listen on port 80 (localhost) and then connect to myserverip on port 443. I get the login screen but can go no further after I Enter my credentials and Click OK. Checking the stunnel log screen it shows that a connect has been made? I don't understand why It won't connect any further?? OWA is on Exchange Server 2003 and I am using Windows XP SP2. I have also updated to Open SSL latest version. Thanks in Advance. Dan

1 0

Stunnel and Outlook Web Access Problems
by Dan Vespa 19 Nov '07

19 Nov '07

I am trying to use the latest version of stunnel to connect to OWA. I set stunnel to listen on port 80 (localhost) and then connect to myserverip on port 443. I get the login screen but can go no further after I Enter my credentials and Click OK. Checking the stunnel log screen it shows that a connect has been made? I don't understand why It won't connect any further?? OWA is on Exchange Server 2003 and I am using Windows XP SP2. I have also updated to Open SSL latest version. Thanks in Advance. Dan

2 1

NFS over stunnel
by Andy Wettstein 19 Nov '07

19 Nov '07

Hi, I wrote a document about how I am running NFS over stunnel. Using some firewall rules I was able to eliminate most of the complications for using secure NFS. It could probably use more detailed explanations, but the scripts I am using are all there. It still kind of in test phase, but I am currently mounting my home directory using stunnel. I haven't really noticed much of a speed problem in day to day activities. https://w3.physics.uiuc.edu/physwiki/doku.php?id=pcs:unix:nfs_over_stunnel Please send any questions you might have. -- andy wettstein unix administrator department of physics university of illinois at urbana-champaign

2 2

receive error
by Protect Girl 14 Nov '07

14 Nov '07

Hi When run stunnel.exe, I was receive "the procedure entry point SSL_CTX_xet_info_callback could not be located in the dynamic library libssl32.dll" error.

2 1

can't get client address
by Draeath Spamtrap 14 Nov '07

14 Nov '07

Stunnel appears to be working fine, but when started it spams my syslog about 5 times with the following message: "stunnel: warning: can't get client address: Bad file descriptor" There are no other messages from stunnel. I'm not sure where it's coming from... It looks like it is running without command line arguments. My init script has the following snippets that make me think that: > PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin > PREFIX=/usr/local > DAEMON=/usr/local/bin/stunnel > PIDFILE=/var/lib/stunnel/stunnel.pid > > test -f $DAEMON || exit 0 > > case "$1" in > start) > echo -n "Starting universal SSL tunnel: stunnel" > $DAEMON || echo -n " failed" > echo "." > ;; > {TRIMMED REST OF FILE} My stunnel version info: > stunnel 4.21 on i686-pc-linux-gnu with OpenSSL 0.9.8e 23 Feb 2007 > Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP > > Global options > debug = 5 > pid = /usr/local/var/run/stunnel/stunnel.pid > RNDbytes = 64 > RNDfile = /dev/urandom > RNDoverwrite = yes > Service-level options > cert = /usr/local/etc/stunnel/stunnel.pem > ciphers = ALL:!ADH:+RC4:@STRENGTH > key = /usr/local/etc/stunnel/stunnel.pem > session = 300 seconds > sslVersion = SSLv3 for client, all for server > TIMEOUTbusy = 300 seconds > TIMEOUTclose = 60 seconds > TIMEOUTconnect = 10 seconds > TIMEOUTidle = 43200 seconds > verify = none My stunnel config file: > cert = /usr/local/etc/stunnel/stunnel.pem > key = /usr/local/etc/stunnel/stunnel.pem > sslVersion = SSLv3 > chroot = /var/lib/stunnel/ > setuid = 99 > setgid = 99 > pid = /stunnel.pid > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > client = yes > [nntps] > accept = 127.0.0.1:119 > connect = secure.news.easynews.com:563 Again, it is all working, and aside from my TCP_NODELAY flags, everything in my conf would be a show-stopper if it was invalid or not working... *of note: I had to set the uid/gid number. for some reason it couldn't resolve the gid/uid on it's own. My /etc/passwd is world-readable (with hashes in shadow) as is standard. I'm not sure I should even be worried about this...

2 2

Reporting bugs in stunnel 4.21
by Michal Trojnara 10 Nov '07

10 Nov '07

Dear Users, Please test the new stunnel 4.21 and report your issues on the Bug Tracking System (http://stunnel.mirt.net/bts.html) or email the information to the stunnel-users mailing list. This is the easiest way to help stunnel working fine on your various systems. I was developing it for quite a long time and I likely failed to fix all the bugs. I'm especially interested in all compilation or runtime issues with the new FIPS and libwrap code. By the time you can find a patch for setuid/setgid+chroot problem here: ftp://stunnel.mirt.net/stunnel/setuid.patch Best regards, Mike

2 2

How is it working?
by fuzzy_4711 07 Nov '07

07 Nov '07

Hello list. After a few tries, my stunnel configuration is working well. I am using it to tunnel my vnc connections to my winXP box. Now I have a question about how the software is working. In the past, when I was using VNC at port 5900 and I did a telnet to that box with port 5900, VNC was answering with something like 003005 which was the VNC protocol version the server was able to communicate. Now because of the tunneling effect, my vnc server still listens at 127.0.0.1:5900 but is expecting ssled connections at xxx.xxx.xxx.xxx:9999. When I do a telnet at xxx.xxx.xxx.xxx at port 9999 my box is answering something like: Connected to xxx.xxx.xxx.xxx Escape character is ... Now if enter something like "test" the telnet window shows me that the connection is closed by foreign host (means: my xp box). Lets assume, someone is trying to hack my computer and doing a port scan. She/he will find out for sure, that my port 9999 is opened. Usually the server listening behind the port is sending something the attacker could use to point to the software running behind the port. In this case, as far as I can see nothing is sent to give a hint that stunnel is waiting there to route my connection attempt to 127.0.0.1:5900. Is it right, that this is the magic - for sure besides encryption and all the algorithms necessary to do the port forwarding - stunnel provides? I mean as long as an attacker doesnt know what is hiding behind the port he/she also doesnt know how to attack or how to get through. Is that conclusion right? Please tell me, if my conclusions are wrong or if I got something wrong. Stefan

3 2

PATCH: Support PostgreSQL SSL
by Marko Kreen 06 Nov '07

06 Nov '07

PostgreSQL protocol has small handshake on connecting before SSL can be initialized. But that means that Stunnel cannot be used by default as it expects the SSL initialization to be the first thing when getting connection. Attached patch adds PostgreSQL protocol support to Stunnel, so that Stunnel can be used to provide SSL when either client or server is not able to do it natively. SSL handshake is described here: http://www.postgresql.org/docs/8.3/static/protocol-flow.html#AEN73013 Patch is tested and works, hopefully it can be included in main Stunnel source. -- marko

1 0

STUNNEL Licensing
by Menon, Sreejith Mohan 06 Nov '07

06 Nov '07

Dear Users, Is there any special licensing required to use STUNNEL ? In case, it is not required from STUNNEL , is there anything to be done on the OPENSSL side ? Thanks and best regards, Sreejith Mohan Menon

1 0

Re: [stunnel-users] almost working
by Tom Allison 05 Nov '07

05 Nov '07

Luke Deller wrote: > Hi Tom, > >> OK, this has been nothing but a bust. > > I've just read this thread, and it looks to me that you have stunnel configured incorrectly for use with inetd. > > Did you see this documentation? > http://www.stunnel.org/faq/stunnel.html#inetd_mode > >> the error keeps saying that port 993 is already in use. > > This is because you have configured stunnel to bind to port 993 using an "accept" option in the configuration file. When using stunnel with inetd, stunnel should not bind to any port; it is inetd which binds to a port. When a client connects to inetd, then inetd will launch stunnel with stunnel's stdin and stdout streams hooked up to the connected socket. > Made some progress. Got past the SSL version. Fixed my firewall to permit IMAPS sessions. confirmed my firewall settings by changing to IMAP and running directly against the imap server. Changed firewall settings from imap to imaps. Keep timing out. Same problem for Mac Mail application (which isn't great) and Debian's Icedove. This is as far as I can get. 2007.11.05 05:43:56 LOG7[27692:47518144360528]: Connection from 65.29.101.30:49841 permitted by libwrap 2007.11.05 05:43:56 LOG5[27692:47518144360528]: inetd connected from 65.29.101.30:49841 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): before/accept initialization 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): SSLv3 read client hello A 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): SSLv3 write server hello A 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): SSLv3 write certificate A 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): SSLv3 write server done A 2007.11.05 05:43:56 LOG7[27692:47518144360528]: SSL state (accept): SSLv3 flush data

1 0

Re: [stunnel-users] Using stunnel for RDP / Proxy / Firewall
by Frank Garber 30 Oct '07

30 Oct '07

I tried using the port you suggested and got the same result. I'm able to verify my firewall is letting the traffic through and that my ISP is not blocking the port by using www.canyouseeme.org . Again, all my settings work when I'm not going through the corporate firewall. Can you send me your whole config file for both your client and server sides? I'm wondering if it has to do with my certificate settings. Thanks, Frank ----- Original Message ---- From: Carter Browne <xxxx> To: garberfc <xxxx> Sent: Monday, October 22, 2007 8:07:11 AM Subject: Re: [stunnel-users] Using stunnel for RDP / Proxy / Firewall I do this all the time. The way I do it is to connect locally to RDP on a non-stardard port. In the RDP dialog box, I have 127.0.0.10:12121, then in stunnel on the local side is: [xxx-rdp] accept = 127.0.0.10:12121 connect = server:12122 client = yes on the remote side is [rdp-incoming] accept = 12122 connect = 3389 client = no. Normally RDP listens for any connection to port 3389, so I found it was easiest to get to to work by moving off that port. Note that you have to open port 12122 in the firewall on the remote side. On the other hand, you can close 3389 on the remote side which takes away an obvious port for hackers. Carter garberfc wrote: > Hi All > > I'm a relative newbie to Stunnel, and am trying to set up a tunnel so I can > Remote Desktop from work to my PC/server at home. > > I'm using versions 4.20 of the Windows binaries. > > I've tested the configuration and it works from home using a laptop that is > going through my firewall > when I enter my domain home (so my firewall is set up correctly). I tried a > variety of common ports and got the same response every time. I had to use > the 127.0.0.2 on the client because Remote Desktop didn't want me connecting > to myself... > > When I try if from work I get a dialog box: > The client could not establish a connection to the remote computer. > The most likely causes for this error are: > 1) Remote connections might not be enabled at the remote computer. > 2)The maximum number of connections was exceeded at the remote computer. > 3) A network error occurred while establishing the connection. > > My config is as follows: > > #Client > ;cert = stunnel.pem > ;key = stunnel.pem > > ; Some performance tunings > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > > ; Some debugging stuff useful for troubleshooting > debug = 7 > output = stunnel.log > > ; Use it for client mode > client = yes > > ; Service-level configuration > [https-RDT] > accept = 127.0.0.2:3389 > connect = xx.xx.xx.xx:1494 > > > #Server > ; Some performance tunings > socket = l:TCP_NODELAY=1 > socket = r:TCP_NODELAY=1 > > ; Some debugging stuff useful for troubleshooting > debug = 7 > output = stunnel.log > > ; Use it for client mode > client = no > > ; Service-level configuration > [https-RDT] > accept = 1494 > connect = localhost:3389 > > > Is there something I need to do to traverse this proxy? Any help would be > greatly appreciated! > >

3 4

Re: [stunnel-users] almost working
by Tom Allison 30 Oct '07

30 Oct '07

On Oct 29, 2007, at 9:39 PM, Luke Deller wrote: > > Hi Tom, > >> OK, this has been nothing but a bust. > > I've just read this thread, and it looks to me that you have > stunnel configured incorrectly for use with inetd. > > Did you see this documentation? > http://www.stunnel.org/faq/stunnel.html#inetd_mode > >> the error keeps saying that port 993 is already in use. > > This is because you have configured stunnel to bind to port 993 > using an "accept" option in the configuration file. When using > stunnel with inetd, stunnel should not bind to any port; it is > inetd which binds to a port. When a client connects to inetd, then > inetd will launch stunnel with stunnel's stdin and stdout streams > hooked up to the connected socket. Amazing how much progess can be made in a minute. Thank you very much. I did miss that line in the docs. I've moved on to a slightly different error, but it's probably going to be easier for me to sort this one out. I have to go to out, so I'll read up on it and try it when I get back. I'm going to guess I have a ssl version/imap version issue to sort out. I'm trying to connect from a macbook to this server. Thank You!!! 2007.10.30 07:07:59 LOG5[22402:47915096707152]: stunnel 4.18 on x86_64-pc-linux-gnu with OpenSSL 0.9.8c 05 Sep 2006 2007.10.30 07:07:59 LOG5[22402:47915096707152]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP 2007.10.30 07:07:59 LOG7[22402:47915096707152]: inetd started 2007.10.30 07:07:59 LOG7[22402:47915096707152]: FD 0 in non-blocking mode 2007.10.30 07:07:59 LOG7[22402:47915096707152]: FD 1 in non-blocking mode 2007.10.30 07:07:59 LOG7[22402:47915096707152]: TCP_NODELAY option set on local socket 2007.10.30 07:07:59 LOG7[22402:47915096707152]: FD 4 in non-blocking mode 2007.10.30 07:07:59 LOG7[22402:47915096707152]: FD 5 in non-blocking mode 2007.10.30 07:07:59 LOG7[22402:47915096707152]: Connection from 65.29.101.30:53036 permitted by libwrap 2007.10.30 07:07:59 LOG5[22402:47915096707152]: inetd connected from 65.29.101.30:53036 2007.10.30 07:07:59 LOG7[22402:47915096707152]: SSL state (accept): before/accept initialization 2007.10.30 07:07:59 LOG3[22402:47915096707152]: SSL_accept: 1408F10B: error:1408F10B:SSL routines:SSL3_GET_RECORD:wrong version number 2007.10.30 07:07:59 LOG5[22402:47915096707152]: Connection reset: 0 bytes sent to SSL, 0 bytes sent to socket 2007.10.30 07:07:59 LOG7[22402:47915096707152]: inetd finished (0 left)

1 0

almost working
by Tom Allison 30 Oct '07

30 Oct '07

I'm trying to tunnel imap but I'm not getting past this: 2007.10.08 05:56:28 LOG5[5355:47656983060560]: stunnel 4.18 on x86_64-pc-linux-gnu with OpenSSL 0.9.8c 05 Sep 2006 2007.10.08 05:56:28 LOG5[5355:47656983060560]: Threading:PTHREAD SSL:ENGINE Sockets:POLL,IPv6 Auth:LIBWRAP 2007.10.08 05:56:28 LOG6[5355:47656983060560]: file ulimit = 1024 (can be changed with 'ulimit -n') 2007.10.08 05:56:28 LOG6[5355:47656983060560]: poll() used - no FD_SETSIZE limit for file descriptors 2007.10.08 05:56:28 LOG5[5355:47656983060560]: 500 clients allowed 2007.10.08 05:56:28 LOG7[5355:47656983060560]: FD 4 in non-blocking mode 2007.10.08 05:56:28 LOG7[5355:47656983060560]: FD 5 in non-blocking mode 2007.10.08 05:56:28 LOG7[5355:47656983060560]: FD 6 in non-blocking mode 2007.10.08 05:56:28 LOG7[5355:47656983060560]: SO_REUSEADDR option set on accept socket 2007.10.08 05:56:28 LOG3[5355:47656983060560]: Error binding imaps to 0.0.0.0:993 2007.10.08 05:56:28 LOG3[5355:47656983060560]: bind: Address already in use (98) I can't find where to begin. I set SO_REUSEADDR and it didn't do anything different. How do I test this stuff sanely?

4 8

stunnel 4.21 released
by Michal Trojnara 27 Oct '07

27 Oct '07

Dear Users, The new version is available for download on: ftp://stunnel.mirt.net/stunnel/ Version 4.21, 2007.10.27, urgency: LOW/EXPERIMENTAL: * New features sponsored by Open-Source Software Institute - Initial FIPS 140-2 support (see INSTALL.FIPS for details). Win32 platform is not currently supported. * New features - Experimental fast support for non-MT-safe libwrap is provided with pre-spawned processes. - Stunnel binary moved from /usr/local/sbin to /usr/local/bin in order to meet FHS and LSB requirements. Please delete the /usr/local/sbin/stunnel when upgrading. - Added code to disallow compiling stunnel with pthreads when OpenSSL is compiled without threads support. - Win32 DLLs for OpenSSL 0.9.8g. - Minor manual update. - TODO file updated. * Bugfixes - Dynamic locking callbacks added (needed by some engines to work). - AC_ARG_ENABLE fixed in configure.am to accept yes/no arguments. - On some systems libwrap requires yp_get_default_domain from libnsl, additional checking was added. - Sending a list of trusted CAs for the client to choose the right certificate restored. - Some compatibility issues with NTLM authentication fixed. - Taskbar icon (unless there is a config file parsing error) and "Save As" disabled in the service mode for local Win32 security (it's much like Yeti -- some people claim they have seen it). sha1 hash for stunnel-4.21.tar.gz file: 7785c45167d902aa728b839adee02a8cc056d86a Best regards, Mike

1 0

Re: [stunnel-users] stunnel-users Digest, Vol 39, Issue 16
by Frank Garber 24 Oct '07

24 Oct '07

I tested with my firewall with both PCs at home. I can make it work with the port open and when I close the port it fails. it's only when I try to connect from behind the corporate firewall that it fails. I've tried several other port numbers and always get the same results. Can you send me your entire client and server config files so I can see the settings? Thanks, Frank ----- Original Message ---- From: "stunnel-users-request(a)mirt.net" <stunnel-users-request(a)mirt.net> To: stunnel-users(a)mirt.net Sent: Tuesday, October 23, 2007 6:00:06 AM Subject: stunnel-users Digest, Vol 39, Issue 16 Send stunnel-users mailing list submissions to stunnel-users(a)mirt.net To subscribe or unsubscribe via the World Wide Web, visit http://stunnel.mirt.net/mailman/listinfo/stunnel-users or, via email, send a message with subject or body 'help' to stunnel-users-request(a)mirt.net You can reach the person managing the list at stunnel-users-owner(a)mirt.net When replying, please edit your Subject line so it is more specific than "Re: Contents of stunnel-users digest..." Today's Topics: 1. [RE] Using stunnel for RDP / Proxy / Firewall (Algol Tradent) Hello, I think your problem might be related to your configuration firewall/router. Have you enabled port-forwarding for the connection? >From your configuration files I can see you are using port 1494. Check that you are forwarding that port. Also make sure you configure any firewall software running on the machine you want to connect to, in order to allow connections. Best Regards __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com _______________________________________________ stunnel-users mailing list stunnel-users(a)mirt.net http://stunnel.mirt.net/mailman/listinfo/stunnel-users

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

stunnel-users