stunnel-users February 2006

stunnel-users@stunnel.org

16 participants
17 discussions

Client setup
by Ting 21 Feb '06

21 Feb '06

Hello all, Wonder if anybody could advise. I have a client tn3270 application running on Solaris 2.8. Currently, it is using clear text to communicate with tn3270 server. Now, I need to run tn3270 with SSL. If I install stunnel in the Solaris 2.8 box to act as a wrapper. 1. Do I run the stunnel as a client only ? Is the command: stunnel -c -d <port1> -r <tn3270server>:<port2> a. Is port1 the original port where client tn3270 application sends to server ? b. Is port2 the same as port1 ? c. Alternatively, am I supposed to change the client tn3270 application to direct the data to a new port X, then port1 = X, while port2 is the port where the server is listening on. Thank you in advance. Warm Regards, TW --------------------------------- Do you Yahoo!? Yahoo! Movies - Search movie info and celeb profiles and photos.

2 1

Choosing local 'source' IP?
by Don Werve 20 Feb '06

20 Feb '06

Ok, so here's the problem: I've got stunnel handling SSL for a web-based service that is otherwise too stupid to be able to talk https. Unfortunately, I need to set up a redirection rule inside this service that bounces people back to the https:// page if they try to come in over http. Since said service isn't managing its own SSL, I have no solid way of determining which connections are wrappered by stunnel and which connections come in over the wire, and if I bounce stunnel connections, I end up in an infinite loop. Fortunately, all stunnel connections give a REMOTE_IP of the second ethernet adapter of the host system, which is good. My question is, how do I configure stunnel to 'force' this behavior? E.g., I want to explicitly specify in stunnel.conf 'Connect to the service for which you are acting as a wrapper from *this* IP/interface'. I've tried the 'local = ' option and setting socket flags for 'l' and 'r', and while they don't break anything, I can't use them to force a connection from any other adapter, so I don't think they are doing what I need. This wouldn't be an issue, either, except I don't know how stunnel determines this, and I don't want to (at some point in the future) have things 'magically change' (read: break horribly). Thanks-in-advance! -- Don Werve <donw(a)iradeon.net> Chief Systems Administrator / Systems Architect

2 1

RE: question about stunnel setup
by Anthony Cicalla 15 Feb '06

15 Feb '06

Encrypting traffic to a remote syslog-ng server including SSL peer authentication 1. Install stunnel and syslog-ng on all machines. 2. Create certificates for all machines. On RedHat 9 and similar machines, you can do the following as root: # cd /etc/pki/tls/certs # make syslog-ng-server.pem # make syslog-ng-client.pem 3. Place copies of syslog-ng-server.pem on all machines in /etc/stunnel with one important alteration. The clients only need the certificate section of syslog-ng-server.pem. In other words, remove the private key section from syslog-ng-server.pem on all clients. Place every client's syslog-ng-client.pem in /etc/stunnel. For server, create a special syslog-ng-client.pem containing the certificate sections for all clients and place in /etc/stunnel. In other words, remove the private key sections from all syslog-ng-client.pem files and concatenate what is left to create server's special syslog-ng-client.pem. Ok my question is from above it says to create one syslog-ng-client.pem file per client. 1)Do the names for each of these need to remain syslog-ng-client.pem or can they have names that reflect the host that they are on? 2) if they all need to keep that name do just move the client file to one client/host and then delete it from the server then move on to create the next?

2 1

stunnel and syslog
by Bart Somers 10 Feb '06

10 Feb '06

Hi to all, We recently ran in to a problem using stunnel on our AIX 5.2 machines. This is our situation: we are running stunnel as a client to protect http (and a proprietary protocol) traffic between our company and a customer. We are using client-athentication to grant access to the server and on our client-side we validate the server-certificates. This all works fine. Due to new security-restrictions, we are moving all our private-key material into HSM-boxes, in our case these are nCipher nShield HSM's. According to the documantion on there website ( http://active.ncipher.com ) we builded openssl nCipher-enabled. Based on this openssl-version, we builded stunnel. All the build-options are shown below in this message. The compiler used is the IBM C compiler. According to the manuals from stunnel we needed to put in an extra line to the config-file which shows "engine = chil" (chil is the Cryptographic Hardware Interface Library, the way the nCipher HSM's are known to openssl) After all this, we created key's in the hsm, using the nCipher tools and loaded them into stunnel (using this method you get a "pseudo-pem-file", which is actually a reference to the hardware-key). Fire-up stunnel using the commands from nCipher to preload the key's and we are all good to go! When this all seems to work very stable after a few months, with thousands of connections and ditto megabytes transfered, we wanted to have all the logging to our syslogserver (it runs syslog-ng) and this is the place where my problem hits in: When commenting-out the option "output = /home/bart/stl/var/stunnel.log" and restarting stunnel, stunnel dies after the first or second connection, leaving no logging, corefiles or whatever behind. Using plain key's (so no HSM, but with the same build which is aware of the HSM), syslog is no problem. Using HSM-key's with plaintext logfiles (no syslog) works also excellent. As i don't have any ideas anymore where to search, i want to ask you if somebody ever have seen this behaviour or knows where it could be caused by. (And maybe how to solve it....) Thanks in advance! Best regards, Bart Somers. ############################# # Build- and config-options # ############################# bart@server1:./stunnel -version stunnel 4.11 on powerpc-ibm-aix5.2.0.0 UCONTEXT+POLL+IPv4 with OpenSSL 0.9.8 05 Jul 2005 Global options cert = home/bart/stl/etc/stunnel/stunnel.pem ciphers = ALL:!ADH:+RC4:@STRENGTH debug = 5 key = /home/bart/stl/etc/stunnel/stunnel.pem pid = /home/bart/stl/var/run/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes session = 300 seconds verify = none Service-level options TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds bart@server1:./openssl version OpenSSL 0.9.8 05 Jul 2005 compile-options: Openssl: $ export LDFLAGS="-L/opt/nfast/toolkits/hwcrhk" $ ./config --prefix=/home/bart/stl –DNFast –ldl make and make install Stunnel: $ export CPPFLAGS="-I/home/bart/stl/include" $ export LDFLAGS="-L/opt/nfast/toolkits/hwcrhk" $ ./configure –-prefix=/home/bart/stl –-with-ssl=/home/bart/stl -–disable-libwrap make and make install Config-file: Include engine = chil Start-up: $ with-nfast –f /home/bart/stl/tmp/ff /home/bart/stl/sbin/stunnel /home/bart/stl/etc/stunnel/stunnel.conf Relevant information from the config-file: engine = chil cert = /home/bart/stl/etc/stunnel/my-key.pem key = /home/bart/stl/etc/stunnel/my-key-hsm.pem ;(no chroot options used) pid = /prj/stl_004/tmp/test/stunnel-client.pid debug = 7 ; output = /home/bart/stl/var/stunnel.log # when in comment, problems! client = yes [test] accept = 9400 connect = 10.10.10.10:443 All other values are the default from the example.

2 4

transparent mode in linux
by Jesse Small 10 Feb '06

10 Feb '06

Hello, I am having trouble with the 'transparent = yes' option in my configuration file. Basically the goal is to see the IP address of the client connecting to our server, but I am getting this error in the log file: bind transparent: Cannot assign requested address (99) Failed to initialize remote connection After searching through the stunnel archives, I saw that this option (-T back then) was supported only up to the Linux 2.2 kernel, but didn't work on the 2.4 kernel. All the messages I read through were a few years old. I am using Fedora Core 4 with a 2.6 kernel, and wanted to see if there were any updates and if it was now possible to see the client's IP address. Otherwise all incoming connections appear as if from localhost. We are using stunnel v4.14. Thanks for your help, Jesse

2 1

stunnel closing connections with RST ?
by sergei 09 Feb '06

09 Feb '06

I'm setting up load-balanced service requiring stunnel with SSL on its front end. There are two hosts behind load balancer running stunnel . Service behind stunnel does not speak SSL. Every few minutes load balancer checks if those stunnels are still alive by opening tcp connection to stunnel listening port. Now problem is that stunnel closes those test connections with RST and load balancer takes it as that host is dead. This is without "client = yes" option because service does not speak SSL load-balancer -> stunnel-host TCP D=1234 S=33007 Syn stunnel-host -> load-balancer TCP D=33007 S=1234 Syn Ack load-balancer -> stunnel-host TCP D=1234 S=33007 Ack load-balancer -> stunnel-host TCP D=1234 S=33007 Fin Ack stunnel-host -> load-balancer TCP D=33007 S=1234 Ack stunnel-host -> load-balancer TCP D=33007 S=1234 Rst On the other hand with client=yes everything works fine load-balancer -> stunnel-host TCP D=123 S=33010 Syn stunnel-host -> load-balancer TCP D=33010 S=123 Syn Ack load-balancer -> stunnel-host TCP D=123 S=33010 Ack load-balancer -> stunnel-host TCP D=123 S=33010 Fin Ack stunnel-host -> load-balancer TCP D=33010 S=123 Fin Ack load-balancer -> stunnel-host TCP D=123 S=33010 Ack Is there any way to make stunnel without "client = yes" close connection "normal way" with FIN instead of RST ? stunnel is last version, load balancer is some older F5 BigIP thanks sergei

3 7

certificate root chain
by Olivier twist 09 Feb '06

09 Feb '06

Hello, I've already sent a message for my problem but no answer. I have a server certificate signed by GlobalSign. I don't want to use client certificate. But if I don't put the certification chain on the CAFILE of stunnel and don't set verify at 1, stunnel doesn't check the server certification chain and the server certificate appears broken on client side !!! I've post this problem on the stunnel mailing list but you tell me that if I don't use client certificate I don't have to set verify at 1. But it doesn't work, and why GlobalSign and others explain how to install server certificatation chain on servers like apache mod ssl?(see http://support.globalsign.net/en/serversign/apachemodssl.cfm) when I read this help file I suppose that the ssl protocol on server side makes a check of server certificate, and that's the reason why the certificate chain appears broken or not on client side. My current ugly solution is to set verify at 1, in this case, on client side, the certificate appears good and not broken but... a dialog box appears and ask for client certificate and some plugin like flash doesn't support that. i use stunnel 4.14 stunnel.conf: cert = c:\certif\inTest.crt key = c:\certif\inTest.key socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 verify = 1 CAfile = c:\certif\ca.pem ;client = yes [https] accept = 443 connect = 127.0.0.1:901 TIMEOUTclose = 0 [rtmps] accept = 80 connect = 127.0.0.1:900 TIMEOUTclose = 0 Could anybody gives me a support? Thx Oliver

3 3

Re: [stunnel-users] certificate root chain
by Hans Werner Strube 09 Feb '06

09 Feb '06

Olivier twist wrote: > I have a server certificate signed by GlobalSign. I don't want to use client > certificate. > But if I don't put the certification chain on the CAFILE of stunnel and > don't set verify at 1, stunnel doesn't check the server certification chain > and the server certificate appears broken on client side !!! ... > cert = c:\certif\inTest.crt > key = c:\certif\inTest.key ... > CAfile = c:\certif\ca.pem AFAIK the whole certificate chain from your server certificate up to the CA certificate should be in inTest.crt (simply concatenate the PEM files). The CAfile would be needed for client verification only.

1 0

RE: [stunnel-users] stunnel help
by Anthony Cicalla 07 Feb '06

07 Feb '06

I sent an email last week for help with a stunnel setup. I am attempting to send syslog messages via a snare agent through stunnel to a fedora core 4 box with syslog-ng installed. I have been able to get one machine to successfully log to the fedora box out of 4. I set these boxes up the same way and I am confused as to why one is working and the others aren't . This is my stunnel.conf file for the clients cert = C:\bin\stunnel\syslog-ng-server.pem CAfile = c:\bin\stunnel\syslog-ng-client.pem verify = 3 [5140] accept = 127.0.0.1:514 connect = 192.168.185.60:5140 This is my stunnel.conf file for the server cert = /etc/stunnel/syslog-ng-server.pem CAfile = /etc/stunnel/syslog-ng-client.pem verify = 3 [5140] accept = 192.168.185.60:5140 connect = 127.0.0.1:514 This is my syslog-ng.conf file on the server options { sync (0); time_reopen (10); log_fifo_size (1000); long_hostnames (off); use_dns (no); use_fqdn (no); create_dirs (no); keep_hostname (yes); }; source s_sys { file ("/proc/kmsg" log_prefix("kernel: ")); unix-stream ("/dev/log"); internal(); udp(ip(0.0.0.0) port(514)); tcp(ip(0.0.0.0) port(514)); }; destination d_cons { file("/dev/console"); }; destination d_mesg { file("/var/log/messages"); }; destination d_auth { file("/var/log/secure"); }; The snare agents are configured to send logs to 127.0.0.1 514. Any thoughts or idea's would be great. sincerely, Anthony -----Original Message----- From: Michal Trojnara [mailto:Michal.Trojnara@mobi-com.net] Sent: Tuesday, February 07, 2006 1:03 AM To: stunnel-users(a)mirt.net Subject: RE: [stunnel-users] stunnel closing connections with RST ? sergei wrote: > Is there any way to make stunnel without "client = yes" > close connection "normal way" with FIN instead of RST ? Stunnel resets connections for a reason. Probably it was reset by the other peer. Check your stunnel log files for details. Best regards, Mike _______________________________________________ stunnel-users mailing list stunnel-users(a)mirt.net http://stunnel.mirt.net/mailman/listinfo/stunnel-users

2 1

Multiple Instances of Same Application, Different Ports
by Steve Ricketts 06 Feb '06

06 Feb '06

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 February 2006