stunnel-users

stunnel-users@stunnel.org

6 participants
3294 discussions

Re: [stunnel-users] certificate verify failed
by Aaron Haywood 05 May '14

05 May '14

Roman, I was wondering if you changed your verify from level 3 to level 2, and if this resolved your peer disconnection error with stunnel. We're experiencing a similar problem but was wondering if anyone out there has resolved this issue already; and if so..how? :) Respectfully, Aaron Haywood | Systems Administrator [cid:image002.png@01CEDA11.E935A7F0] O: 858.614.1547 | M: 858.699.8302 | www.sdhealthconnect.org CONNECT WITH US: Facebook<https://www.facebook.com/SDHealthConnect?filter=1> | Twitter<https://twitter.com/SDHealthConnect> | LinkedIn<http://www.linkedin.com/company/san-diego-health-connect?trk=company_logo> | Google+<https://plus.google.com/u/0/b/112077458420951237937/112077458420951237937/p…> | YouTube<http://www.youtube.com/SDHealthConnect>

1 0

New stunnel how-to
by Patrick J. 30 Apr '14

30 Apr '14

Hi all, I've spent some time to write a new stunnel how-to. - Install and configure stunnel on a Raspberry Pi - Run stunnel in a chroot jail - Limit allowed ciphers and force Forward Secrecy - Restrict access to the stunnel server (verify = 3) - Disguise SSH as HTTPS traffic http://stunnel.qnea.net/ Regards from Germany, Patrick

1 1

Error 126 compiling on Mac OS X 10.7.5
by James Brown 29 Apr '14

29 Apr '14

I’ve just tried to install version 5.01 and encountered an error. I had no problems installing 5.01b1 previously. I just thought I’d install the release version. System is running Mac OS X 10.7.5. make ends with a warning, and make install fails: ld: warning: directory not found for option '-L/usr//lib64' Making all in doc make[1]: Nothing to be done for `all'. Making all in tools make[1]: Nothing to be done for `all'. make[1]: Nothing to be done for `all-am'. mail:stunnel-5.01 2 jlbrown$ sudo make install Password: Making install in src ../auto/install-sh -c -d '/usr/local/bin' /bin/sh ../libtool --mode=install /usr/bin/install -c stunnel '/usr/local/bin' libtool: install: /usr/bin/install -c stunnel /usr/local/bin/stunnel ../auto/install-sh -c -d '/usr/local/bin' /usr/bin/install -c stunnel3 '/usr/local/bin' ../auto/install-sh -c -d '/usr/local/lib/stunnel' /bin/sh ../libtool --mode=install /usr/bin/install -c libstunnel.la '/usr/local/lib/stunnel' libtool: install: /usr/bin/install -c .libs/libstunnel.dylib /usr/local/lib/stunnel/libstunnel.dylib libtool: install: /usr/bin/install -c .libs/libstunnel.lai /usr/local/lib/stunnel/libstunnel.la /Users/jlbrown/Downloads/stunnel-5.01: /Users/jlbrown/Downloads/stunnel-5.01: is a directory make[2]: *** [install-pkglibLTLIBRARIES] Error 126 make[1]: *** [install-am] Error 2 make: *** [install-recursive] Error 1 I just used ./configure with no arguments. Any suggestions? Thanks, James.

1 0

Verify=2 and Verification Depth...
by Matt Wise 28 Apr '14

28 Apr '14

It was my understanding that when you have an Stunnel Server configured with 'verify=2', that the client that connects must have a certificate signed by the same CA/SubCA combination that the server does. So for example: - My_Root_Ca (private CA) - Some_Random_Cert.pem - Stunnel_Sub_Ca: - Server.pem - Client.pem - Postgres_Sub_Ca: - Server.pem - postgres_user.pem With the above structure in place (and the stunnel server using Stunnel_Sub_Ca/Server.pem) if someone tried to connect in with the Stunnel_Sub_Ca/Client.pem cert, it would work... but if they tried to connect in with Postgres_Sub_Ca/Server.pem, it wouldn't. Unfortunately we're not seeing that behavior... we're seeing a behavior where *every* cert signed by the overall Root CA is validated. We're able to connect in using Some_Random_Cert.pem, Postgres_Sub_Ca/Server.pem and Postgres_Sub_Ca/postgres_user.pem. This feels wrong ... what am I missing? (We're using Stunnel 4.55 btw) Matt Wise Sr. Systems Architect Nextdoor.com

1 1

ephemeral port limit
by Frank Liu 28 Apr '14

28 Apr '14

Hi, I am trying to use stunnel to add ssl support to my app. My app and linux server are tuned to accept 200k+ connections without a problem. When adding stunnel on the same server, the connection stops when it reaches 64k due to stunnel uses ephemeral ports to connect to my app on the localhost. Any suggestions how to get around this limit? One thing I am thinking of is to add 10 virtual interfaces on loopback: lo0:2, lo0:3, etc with ip 127.0.0.2, 127.0.0.3 etc. Can stunnel be configured to cycle through a pool of ip for outbound connection to my app listening on 127.0.0.1? That would give 10x64k of ephemeral ports. I see stunnel has "local" to specify outgoing ip, but I am not sure if it supports a list of ip or use multiple times. If so, will it round robin the list? Thanks Frank

2 2

Syringe
by Michal Trojnara 21 Apr '14

21 Apr '14

Hi Guys, I'm glad to announce syringe, a service for testing client-side heartbleed. It allows to easily examine most of the obscure TLS clients (for example embedded devices). The service is available online at https://www.stunnel.org:4433/ Best regards, Mike

2 2

hardware random
by ios dev 20 Apr '14

20 Apr '14

Hello. I am using GOST engine and set it as default for ALL. At the same time i want to use hardware RNG from other engine. How can i do this? -- BR, Dev.

1 0

Problem with round robin on stunnel 5.0 and 5.1
by WNSDEV 14 Apr '14

14 Apr '14

When I use either of these two versions of stunnel the round-robin for "connect" only sends connections to the first host listed on the configuration file. However the round-robin works in Stunnel 4.56,4.55,4.54 and 4.53. Anyone else have this issue? Thank you, Pete Sangas ./stunnel -version stunnel 5.00 on x86_64-unknown-linux-gnu platform Compiled/running with OpenSSL 1.0.1g 7 Apr 2014 Threading:PTHREAD Sockets:POLL,IPv6 SSL:ENGINE,OCSP,FIPS Global options: debug = daemon.notice RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options: ciphers = FIPS (with "fips = yes") ciphers = HIGH:MEDIUM:+3DES:+DH:!aNULL:!SSLv2 (with "fips = no") curve = prime256v1 sessionCacheSize = 1000 sessionCacheTimeout = 300 seconds sslVersion = TLSv1 (with "fips = yes") sslVersion = TLSv1 for client, all for server (with "fips = no") stack = 65536 bytes TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none Config file : CApath = /xxx/certs/trusted CAfile = /yyy/cacert.pem ciphers = RC4-SHA debug = daemon.7 pid = /zzz/stunnel.pid options = NO_SSLv2 foreground = yes setgid = gid1 setuid = uid1 delay = no fips=no [stunnel] TIMEOUTidle=86400 verify = 3 cert = /path/servercert.pem accept = 443 failover=rr connect = 10.1.1.50:23 connect = 10.1.1.60:23

2 2

DH not working, although initialized correctly
by Ovidiu Lixandru 14 Apr '14

14 Apr '14

Hello, I'm trying to enable Forward Secrecy. have an stunnel 4.56 instance compiled with authpriv and xforwardedfor patches, configured as follows: sslVersion = all options = NO_SSLv2 ciphers = ECDHE-RSA-AES256-GCM-SHA384:ECDHE-RSA-AES128-GCM-SHA256:DHE-RSA-AES256-GCM-SHA384:DHE-RSA-AES128-GCM-SHA256:ECDHE-RSA-AES256-SHA384:ECDHE-RSA-AES128-SHA256:ECDHE-RSA-AES256-SHA:ECDHE-RSA-AES128-SHA:DHE-RSA-AES256-SHA256:DHE-RSA-AES128-SHA256:DHE-RSA-AES256-SHA:DHE-RSA-AES128-SHA:ECDHE-RSA-DES-CBC3-SHA:EDH-RSA-DES-CBC3-SHA:AES256-GCM-SHA384:AES128-GCM-SHA256:AES256-SHA256:AES128-SHA256:AES256-SHA:AES128-SHA:DES-CBC3-SHA:HIGH:!aNULL:!eNULL:!EXPORT:!CAMELLIA:!DES:!MD5:!PSK:!RC4 renegotiation = no setuid = nobody setgid = nobody socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 libwrap = no TIMEOUTbusy = 120 xforwardedfor = yes [domain.com] accept = X.X.X.X:443 connect = 127.0.0.1:91 cert = /etc/pki/tls/certs/domain.pem key = /etc/pki/tls/private/domain.key domain.pem contains both the SSL certificate and the DH parameters (generated with "openssl dhparams 2048"). stunnel initializes the DH parameters correctly: 2014.04.14 12:08:18 LOG6[16629:140355580176320]: Initializing service [domain.com] 2014.04.14 12:08:18 LOG7[16629:140355580176320]: Certificate: /etc/pki/tls/certs/domain.pem 2014.04.14 12:08:18 LOG7[16629:140355580176320]: Certificate loaded 2014.04.14 12:08:18 LOG7[16629:140355580176320]: Key file: /etc/pki/tls/private/domain.key 2014.04.14 12:08:18 LOG7[16629:140355580176320]: Private key loaded 2014.04.14 12:08:18 LOG7[16629:140355580176320]: Using DH parameters from /etc/pki/tls/certs/domain.pem 2014.04.14 12:08:18 LOG7[16629:140355580176320]: DH initialized with 2048-bit key 2014.04.14 12:08:18 LOG7[16629:140355580176320]: SSL options set: 0x01000004 When I try to connect using openssl s_client, the connections fails with it unable to negotiate a cipher. On the client side: # openssl s_client -cipher ECDH -tls1 -connect www.domain.com:443 CONNECTED(00000003) 140735113126752:error:14094410:SSL routines:SSL3_READ_BYTES:sslv3 alert handshake failure:s3_pkt.c:1256:SSL alert number 40 140735113126752:error:1409E0E5:SSL routines:SSL3_WRITE_BYTES:ssl handshake failure:s3_pkt.c:596: --- no peer certificate available --- No client certificate CA names sent --- SSL handshake has read 7 bytes and written 0 bytes --- New, (NONE), Cipher is (NONE) Secure Renegotiation IS NOT supported Compression: NONE Expansion: NONE SSL-Session: Protocol : TLSv1 Cipher : 0000 Session-ID: Session-ID-ctx: Master-Key: Key-Arg : None PSK identity: None PSK identity hint: None SRP username: None Start Time: 1397471905 Timeout : 7200 (sec) Verify return code: 0 (ok) --- On the server side: 2014.04.14 11:57:10 LOG5[9647:140038728394496]: Service [domain.com] accepted connection from X.X.X.X:56678 2014.04.14 11:57:10 LOG7[9647:140038728394496]: SSL state (accept): before/accept initialization 2014.04.14 11:57:10 LOG7[9647:140038728394496]: SNI: no virtual services defined 2014.04.14 11:57:10 LOG7[9647:140038728394496]: SSL alert (write): fatal: handshake failure 2014.04.14 11:57:10 LOG3[9647:140038728394496]: SSL_accept: 1408A0C1: error:1408A0C1:SSL routines:SSL3_GET_CLIENT_HELLO:no shared cipher 2014.04.14 11:57:10 LOG5[9647:140038728394496]: Connection reset: 0 byte(s) sent to SSL, 0 byte(s) sent to socket The client and the server have the same openssl version (1.0.1e), with the same supported ciphers. I tried different cipher combinations (i.e. "HIGH:!aNULL:!MD5"), but no lock. FWIW, the initial DH cipher list in the stunnel config file works correctly in an SSL nginx instance and openssl s_client negotiates "ECDHE-RSA-AES256-SHA". Any idea what may be wrong with stunnel? Thanks, Ovidiu

1 0

Re: [stunnel-users] Some troubles with PROXY protocol
by Alexey V. Drozdov 13 Apr '14

13 Apr '14

Hi, Mike! In spite of new changes, it's still work incorrect because of init_remote(c) connect to the address from root config section: protocol(c, PROTOCOL_PRE_CONNECT); init_remote(c); <<<<<<<<<< destination address from root section protocol(c, PROTOCOL_PRE_SSL); init_ssl(c); <<<<<<<<<<< switch to target SNI config section only there protocol(c, PROTOCOL_POST_SSL); <<<<<<<<<<< send PROXY protocol to incorrect address Context switch to target section parameters after init_ssl(c) only. /Alexey V. Drozdov e-mail: anyquist(a)yandex.ru

2 2

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

stunnel-users