- stunnel-users - stunnel.org

Re: [stunnel-users] build static stunnel with openssl
by Peter Pentchev 08 Oct '18

08 Oct '18

On Mon, Oct 08, 2018 at 03:09:07PM +0800, White Little wrote: > Thanks, Eric > > But I did not put it on /usr/local/opensl or /usr/local/stunnel > > > Below is my configuration log > > > [root@Stunnel-Fedora stunnel-5.49]# ./configure --enable-static > --with-ssl=/root/stunnel/openssl-1.1.0i [snip] > configure: **************************************** TLS [snip] > checking for TLS directory... /root/stunnel/openssl-1.1.0i [snip] > > I tested on ubuntu14.04, 16.04 64bit and fedora 32bit, all of them are > them are not include opennssl. What exactly makes you think that this does not include OpenSSL support? The configure script seemed to find the include/openssl/ directory in the directory that you specified; does the build fail somehow later? If it does, show us the logs from the failed build. Or does the compiled executable does not say "Compiled with OpenSSL 1.1.0i"? If it does not, show us the full build log (maybe use a pastebin-like service somewhere, e.g. paste.debian.net) BTW a question: is /root/stunnel/openssl-1.1.0i the place you extracted and compiled OpenSSL, or the place you installed it after building it? If you just built it there, you also need to install it somewhere and point the stunnel configure script at the directory where it is installed, not at the build tree. This might mean that you may need to rebuild it again after specifying a --prefix or something like that (I do not remember exactly how the OpenSSL build works). G'luck, Peter -- Peter Pentchev roam(a){ringlet.net,debian.org,FreeBSD.org} pp(a)storpool.com PGP key: http://people.FreeBSD.org/~roam/roam.key.asc Key fingerprint 2EE7 A7A5 17FC 124C F115 C354 651E EFB0 2527 DF13

1 0

Re: [stunnel-users] build static stunnel with openssl
by Eric Eberhard 06 Oct '18

06 Oct '18

Make openssl and stunnel static. That is what I do and I consider it MUCH smarter than dynamic. I got tired of getting support calls because stunnel stopped working because the user installed a new openssl. Static will always work. Another piece of advice – do NOT put it on /usr/local/opensl or /usr/local/stunnel (or whatever the defaults are). Before building make your OWN directories. Our short name for our product is “met” so we use /usr/local/met/obj and /usr/local/met/bin for the object and binary files, respectively. This ensures nobody will install over you. This is also a real security issue mostly ignored. Most people have heard of SQL injection (where SQL is modified to do what the developer never intended). However, DLL injection (e.g. dynamic library injection) is a real problem. I could easily install an openssl DLL that does exactly what it should do … and sends the same data in clear text (or with my own encryption) to my machine. A thief could simply install one dynamic library and then OWN your data. E From: stunnel-users [mailto:stunnel-users-bounces@stunnel.org] On Behalf Of White Little Sent: Sunday, September 30, 2018 11:33 PM To: stunnel-users(a)stunnel.org Subject: [stunnel-users] build static stunnel with openssl Hi All I try to build a static (portable) stunnel that contain opensll. I know there are some similar questions long ago, but I still fail on how to build it. I try to link the stunnel to latest openssl, so I download the openssl-1.0.2p and make it. And then I type ./configure --enable-static --with-ssl= openssl-1.0.2p_dir And typing make, I found there is a stunnel binary under /src folder, but it still not include the openssl. Do I miss something that I did not notice? Thanks

1 0

server mode support for both ECC and RSA server certificates
by Tom Hood 02 Oct '18

02 Oct '18

Hi, Can one stunnel service in server mode be configured with both ECDSA and RSA server certs? The stunnel man page Certificates section suggests only one can be specified along with its CA chain. Use case is some legacy clients don't support ECDSA, but it would be nice to support ECDSA when a client does support it. It appears that HAProxy supports this <https://www.haproxy.com/blog/serving-ecc-and-rsa-certificates-on-same-ip-wi…>,but I'm hoping there is an equivalent way to do this with stunnel. I'm currently testing with stunnel 5.49 with OpenSSL 1.0.2p on Solaris 11.3, but can rebuild with any version of openssl if that helps. Thanks, -- Tom

1 0

build static stunnel with openssl
by White Little 01 Oct '18

01 Oct '18

Hi All I try to build a static (portable) stunnel that contain opensll. I know there are some similar questions long ago, but I still fail on how to build it. I try to link the stunnel to latest openssl, so I download the openssl-1.0.2p and make it. And then I type *./configure --enable-static --with-ssl= openssl-1.0.2p_dir* And typing make, I found there is a stunnel binary under /src folder, but it still not include the openssl. Do I miss something that I did not notice? Thanks

1 0

Re: [stunnel-users] stunnel log rolling
by Eric Eberhard 17 Sep '18

17 Sep '18

Most O/S will either prevent renaming a file that is in use (or deleting it). Some will – but they will actually continue to write to the old file (and if that file was deleted then it writes to a file you can nexer see). Since I generally use AIX it is the second feature I see. We have systems with over 500 connections via stunnel – it never rests, and rolling the log file is not easy. I can easily provide a code fix for stunnel that we use in out software. It allows daily logs, sized logs, etc. And it rolls properly. The downside is we open the log, write to it, and close it – each time – which is not efficient. Given modern computers I have come to not care. So the open routine either opens (append) a daily file or it looks up the file and gets the size and decides to move it (or not). The second is not perfect … if three other processes have it open at the moment they will finish writing to that log. Who cares? As long as the logging is always opening and closing the “bleed over” is minimized. If people want me to do this, I will – if they can tell me the log routines it would speed me some, preventing from having to search for them. We like and prefer the daily logs – it is clean as you get one per day. As long as one day is never too huge it is great and we have daemons that remove logs over xx number of days (although I could add a simple parameter for that as well and simply kill those files). This is an annoying problem – especially for multi-O/S software – but one I am good at. Eric From: stunnel-users [mailto:stunnel-users-bounces@stunnel.org] On Behalf Of Daniel Trickett Sent: Wednesday, September 12, 2018 11:16 AM To: Tom Hood <tom.w.hood(a)gmail.com> Cc: stunnel-users(a)stunnel.org Subject: Re: [stunnel-users] stunnel log rolling Hi Tom, Is what you refer to? I think the open and re-open only happen when the service is stopped and restarted. It hasn’t rolled over like Apache in my short experience. log = append | overwrite log file handling This option allows you to choose whether the log file (specified with the output option) is appended or overwritten when opened or re-opened. default: append Best regards, Dan Daniel Trickett Head of Database Services | MBS Business Technology BX-TCS-O Oracle ERP Business Services of Merck KGaA, Darmstadt, Germany Planned Absence – MilliporeSigma A business of Merck KGaA, Darmstadt, Germany EMD Millipore Corporation | 80 Ashby Road | Bedford, MA 01730 | USA office 781-533-3017 |cell 978-761-3506 |email <mailto:daniel.trickett@emdmillipore.com> daniel.trickett(a)emdmillipore.com From: Tom Hood <tom.w.hood(a)gmail.com <mailto:tom.w.hood@gmail.com> > Sent: Wednesday, September 12, 2018 1:10 PM To: Daniel Trickett <daniel.trickett(a)emdmillipore.com <mailto:daniel.trickett@emdmillipore.com> > Cc: stunnel-users(a)stunnel.org <mailto:stunnel-users@stunnel.org> Subject: Re: [stunnel-users] stunnel log rolling Hi Daniel, The trick is how to roll the logs without an interruption of service (i.e. without a stunnel restart). I believe stunnel claims to support this, but I think the feature might be broken in 5.49 Thanks, -- Tom On Wed, Sep 12, 2018 at 5:43 AM Daniel Trickett <daniel.trickett(a)emdmillipore.com <mailto:daniel.trickett@emdmillipore.com> > wrote: Tom, Kill the stunnel process. Then mv the log. This will allow stunnel to right to a new log file. Best regards, Dan Daniel Trickett Head of Database Services | MBS Business Technology BX-TCS-O Oracle ERP Business Services of Merck KGaA, Darmstadt, Germany Planned Absence – MilliporeSigma A business of Merck KGaA, Darmstadt, Germany EMD Millipore Corporation | 80 Ashby Road | Bedford, MA 01730 | USA office 781-533-3017 |cell 978-761-3506 |email <mailto:daniel.trickett@emdmillipore.com> daniel.trickett(a)emdmillipore.com From: stunnel-users <stunnel-users-bounces(a)stunnel.org <mailto:stunnel-users-bounces@stunnel.org> > On Behalf Of Tom Hood Sent: Tuesday, September 11, 2018 5:02 PM To: stunnel-users(a)stunnel.org <mailto:stunnel-users@stunnel.org> Subject: [stunnel-users] stunnel log rolling Hi, I'm new to stunnel and it isn't clear to me how the log rolling feature works. I built stunnel 5.49 with gcc 4.2.0 on Solaris 10. I'm running it on Solaris 11.3 SPARC. Using openssl 1.0.2p The config file has disabled syslog and is logging to stunnel.log. Command line is: stunnel stunnel.conf where stunnel.conf contains the following: syslog = no output = stunnel.log debug = 7 [service-exterior] client = no options = NO_SSLv2 options = NO_SSLv3 options = NO_TLSv1 options = NO_TLSv1.1 options = -NO_TLSv1.2 cert = /path/to/stunnel.pem curve = zzz accept = testhost:32100 connect = 127.0.0.1:32200 <http://127.0.0.1:32200> [service-interior] client = yes options = NO_SSLv2 options = NO_SSLv3 accept = 127.0.0.1:32200 <http://127.0.0.1:32200> connect = 127.0.0.1:32100 <http://127.0.0.1:32100> sslVersion = TLSv1 ciphers = zzz TIMEOUTconnect = 60 The log rollowing steps I tried that don't work are: mv stunnel.log stunnel.log.1 kill -USR1 <stunnelpid> The log message "LOG7[main]: Processing SIGNAL_REOPEN_LOG" shows up in stunnel.log.1. However, new client connections to host:32100 do not trigger creation of a new stunnel.log file. In fact, logging stops to stunnel.log.1 as soon as the USR1 is processed. The new client connections work as before, but there isn't any logging. I restarted stunnel and tried the test again with these steps: mv stunnel.log stunnel.log.1 touch stunnel.log kill -USR1 <stunnelpid> That also doesn't work. Please let me know the correct sequence of steps to roll the stunnel.log Thank you, -- Tom This message and any attachment are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, you must not copy this message or attachment or disclose the contents to any other person. If you have received this transmission in error, please notify the sender immediately and delete the message and any attachment from your system. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not accept liability for any omissions or errors in this message which may arise as a result of E-Mail-transmission or for damages resulting from any unauthorized changes of the content of this message and any attachment thereto. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not guarantee that this message is free of viruses and does not accept liability for any damages caused by any virus transmitted therewith. Click http://www.emdgroup.com/emd/imprint/mail_disclaimer.html to access the German, French, Spanish and Portuguese versions of this disclaimer. This message and any attachment are confidential and may be privileged or otherwise protected from disclosure. If you are not the intended recipient, you must not copy this message or attachment or disclose the contents to any other person. If you have received this transmission in error, please notify the sender immediately and delete the message and any attachment from your system. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not accept liability for any omissions or errors in this message which may arise as a result of E-Mail-transmission or for damages resulting from any unauthorized changes of the content of this message and any attachment thereto. Merck KGaA, Darmstadt, Germany and any of its subsidiaries do not guarantee that this message is free of viruses and does not accept liability for any damages caused by any virus transmitted therewith. Click http://www.emdgroup.com/emd/imprint/mail_disclaimer.html to access the German, French, Spanish and Portuguese versions of this disclaimer.

2 1

stunnel log rolling
by Tom Hood 12 Sep '18

12 Sep '18

Hi, I'm new to stunnel and it isn't clear to me how the log rolling feature works. I built stunnel 5.49 with gcc 4.2.0 on Solaris 10. I'm running it on Solaris 11.3 SPARC. Using openssl 1.0.2p The config file has disabled syslog and is logging to stunnel.log. Command line is: stunnel stunnel.conf where stunnel.conf contains the following: syslog = no output = stunnel.log debug = 7 [service-exterior] client = no options = NO_SSLv2 options = NO_SSLv3 options = NO_TLSv1 options = NO_TLSv1.1 options = -NO_TLSv1.2 cert = /path/to/stunnel.pem curve = zzz accept = testhost:32100 connect = 127.0.0.1:32200 [service-interior] client = yes options = NO_SSLv2 options = NO_SSLv3 accept = 127.0.0.1:32200 connect = 127.0.0.1:32100 sslVersion = TLSv1 ciphers = zzz TIMEOUTconnect = 60 The log rollowing steps I tried that don't work are: mv stunnel.log stunnel.log.1 kill -USR1 <stunnelpid> The log message "LOG7[main]: Processing SIGNAL_REOPEN_LOG" shows up in stunnel.log.1. However, new client connections to host:32100 do not trigger creation of a new stunnel.log file. In fact, logging stops to stunnel.log.1 as soon as the USR1 is processed. The new client connections work as before, but there isn't any logging. I restarted stunnel and tried the test again with these steps: mv stunnel.log stunnel.log.1 touch stunnel.log kill -USR1 <stunnelpid> That also doesn't work. Please let me know the correct sequence of steps to roll the stunnel.log Thank you, -- Tom

2 4

stunnel 5.49 released
by Michal Trojnara 03 Sep '18

03 Sep '18

Dear Users, I have released version 5.49 of stunnel. Version 5.49, 2018.09.03, urgency: MEDIUM * New features - Performance optimizations. - Logging of negotiated or resumed TLS session IDs (thx to ANSSI - National Cybersecurity Agency of France). - Merged Debian 10-enabled.patch and 11-killproc.patch (thx to Peter Pentchev). - OpenSSL DLLs updated to version 1.0.2p. - PKCS#11 engine DLL updated to version 0.4.9. * Bugfixes - Fixed a crash in the session persistence implementation. - Fixed syslog identifier after configuration file reload. - Fixed non-interactive "make check" invocations. - Fixed reloading syslog configuration. - stunnel.pem created with SHA-256 instead of SHA-1. - SHA-256 "make check" certificates. Home page: https://www.stunnel.org/ Download: https://www.stunnel.org/downloads.html SHA-256 hashes: 3d6641213a82175c19f23fde1c3d1c841738385289eb7ca1554f4a58b96d955e stunnel-5.49.tar.gz 459bbb212baf0b9821c80e0664c830246ef6e97c7329fb08160e87ff11ae9692 stunnel-5.49-win32-installer.exe 72416c6664106ad815a8da67a525c6593247fc06cbca3b8918ffc87ae92595e8 stunnel-5.49-android.zip Best regards, Mike

1 0

Adding custom HTTP headers to a proxy'ed HTTP request when stunnel is used as an endpoint for a HTTP proxy
by Tony Cheneau 02 Sep '18

02 Sep '18

Good evening folks, Michal was kind enough to review and rewrite/integrate one of my patches for logging session IDs [1] and post the remaining ones on the website [2]! So, I feel I should also contribute to this list a little program I wrote. It is meant to be used with the exec= directive. It is mostly useful when stunnel is used to secure a connexion between a client and a HTTP proxy. A stunnel instance can run locally on the client while a stunnel server runs on the HTTP proxy: HTTP client (proxy'ed) <- > stunnel client <--> stunnel server <-> HTTP proxy Here, the "exec=" program will run on the stunnel server and add some extra headers to each HTTP request. One of this headers is the client's Distinguished Name (from its X.509 certificate). The HTTP proxy (squid for example) can then make some decision based on this HTTP header (authorisation, etc.). It relies heavily on PicoHTTPParser. Regards, Tony Cheneau [1]: to appear in version 5.49, see https://www.stunnel.org/sdf_ChangeLog.html [2]: patches 0002 and 0003 on https://www.stunnel.org/patches.html

1 0

OpenSSL 1.1.1 secure level and also an old problem rearing up
by Peter Pentchev 26 Aug '18

26 Aug '18

Hi, Sorry to be the bearer of a "those OS vendors did something again and now we have to catch up with them... again..." type of news, but, well, the maintainers of the Debian package of OpenSSL upgraded it to a prerelease 1.1.1 version and, in the process, changed the default cipher selection in the openssl.cnf file to 'SECLEVEL=2'. Unfortunately, this means that the server certificate used in the stunnel test suite is not considered strong enough, so the test suite is failing. More details may be found in the description of the Debian bug https://bugs.debian.org/906981 and the error message itself is at https://ci.debian.net/data/autopkgtest/testing/amd64/s/stunnel4/865956/log.… (look for all the lines containing "SSL_CTX_use_certificate:ca md too weak") So my first reaction was to stop running the stunnel test suite and only leave my functional test script; however, I then had another idea and tried to get the stunnel test suite to explicitly specify SECLEVEL=1 using the attached patch. This actually helped a bit - it made most of the tests pass - but then I hit a problem with e.g. 037_failover_prio1: if there is a "ciphers" option in the config file, stunnel eventually dies with an error that I seem to remember having seen before; take a look at this gdb backtrace from stunnel 5.48: (gdb) bt #0 __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:51 #1 0x00007f0a349882f1 in __GI_abort () at abort.c:79 #2 0x000055b1aa3ad645 in fatal_debug (txt=0x55b1aa3c503c "Bad magic", file=0x55b1aa3ca6dd "ssl.c", line=118) at log.c:379 #3 0x000055b1aa3ae066 in get_alloc_list_ptr (ptr=0x7f0a24014520, file=0x55b1aa3ca6dd "ssl.c", line=118) at str.c:429 #4 0x000055b1aa3ae229 in str_detach_debug (ptr=ptr@entry=0x7f0a24014520, file=file@entry=0x55b1aa3ca6dd "ssl.c", line=line@entry=118) at str.c:378 #5 0x000055b1aa3ae884 in str_detach_debug (line=118, file=0x55b1aa3ca6dd "ssl.c", ptr=0x7f0a24014520) at str.c:413 #6 str_free_debug (ptr=0x7f0a24014520, file=0x55b1aa3ca6dd "ssl.c", line=118) at str.c:413 #7 0x00007f0a34f7661a in CRYPTO_free_ex_data () from /usr/lib/x86_64-linux-gnu/libcrypto.so.1.1 #8 0x00007f0a35112a98 in SSL_SESSION_free () from /usr/lib/x86_64-linux-gnu/libssl.so.1.1 #9 0x00007f0a34f7a1bc in OPENSSL_LH_doall_arg () from /usr/lib/x86_64-linux-gnu/libcrypto.so.1.1 #10 0x00007f0a35114177 in SSL_CTX_flush_sessions () from /usr/lib/x86_64-linux-gnu/libssl.so.1.1 #11 0x000055b1aa3c3701 in unbind_ports () at stunnel.c:439 #12 0x000055b1aa3c3799 in main_cleanup () at stunnel.c:224 #13 0x000055b1aa3ad720 in main (argc=<optimized out>, argv=<optimized out>) at ui_unix.c:79 (gdb) (the line numbers may be a bit off from the stock stunnel source, since I also added a setrlimit(RLIMIT_CORE, infinity) patch) So, yeah, what would be the best way forward here? I guess I might create a new certificate specifically for the Debian package's test suite, but I think the double-free problem may also need to be fixed. Let me know if you need any more information about reproducing it or any more data in case it doesn't happen on your end. Thanks for everything you're doing for stunnel, and keep up the great work! G'luck, Peter -- Peter Pentchev roam(a){ringlet.net,debian.org,FreeBSD.org} pp(a)storpool.com PGP key: http://people.FreeBSD.org/~roam/roam.key.asc Key fingerprint 2EE7 A7A5 17FC 124C F115 C354 651E EFB0 2527 DF13

2 2

Log to syslog without process name, after reloading the configuration (sighup)
by Reschke, Sven (VWIF G-TS/P) 22 Aug '18

22 Aug '18

Hi, I've upgraded my stunnel from version 5.44 to 5.45. After reloading the configuration of stunnel (via pkill -sighup stunnel) several times, I've stumbled about missing syslogs. Only the loading of the configuration file was still visible. Then I've noticed, that greping the syslog for stunnel was my mistake. Stunnel doesn't log its process name after reloading the configuration file the second time (see logs below) Can someone verify, or is it just me, or some configuration I'm missing? BTW: I've reproduced the "problem" also for the version 5.48 as you can see from the logs. With version 5.44 stunnel always logged its process name to syslog. Syslog output for greping after "stunnel": Aug 17 13:41:14 <user> stunnel: LOG5[ui]: stunnel 5.48 on armv7l-unknown-linux-gnueabihf platform Aug 17 13:41:14 <user> stunnel: LOG5[ui]: Compiled/running with OpenSSL 1.0.2p 14 Aug 2018 Aug 17 13:41:14 <user> stunnel: LOG5[ui]: Threading:PTHREAD Sockets:POLL,IPv6 TLS:ENGINE,FIPS,OCSP,PSK,SNI Aug 17 13:41:14 <user> stunnel: LOG5[ui]: Reading configuration from file /home/pi/stunnel.conf Aug 17 13:41:14 <user> stunnel: LOG5[ui]: UTF-8 byte order mark not detected Aug 17 13:41:14 <user> stunnel: LOG5[ui]: FIPS mode disabled Aug 17 13:41:14 <user> stunnel: LOG4[ui]: Insecure file permissions on /home/pi/psk.txt Aug 17 13:41:14 <user> stunnel: LOG4[ui]: Insecure file permissions on /home/pi/psk.txt Aug 17 13:41:14 <user> stunnel: LOG5[ui]: Configuration successful ---------------- Reloading the configuration the first time ----------------------- Aug 17 13:41:21 <user> stunnel: LOG5[main]: Reading configuration from file /home/pi/stunnel.conf Aug 17 13:41:21 <user> stunnel: LOG5[main]: UTF-8 byte order mark not detected Aug 17 13:41:21 <user> stunnel: LOG5[main]: FIPS mode disabled Aug 17 13:41:21 <user> stunnel: LOG4[main]: Insecure file permissions on /home/pi/psk.txt Aug 17 13:41:21 <user> stunnel: LOG4[main]: Insecure file permissions on /home/pi/psk.txt Aug 17 13:41:21 <user> stunnel: LOG5[main]: Configuration successful ---------------- Reloading the configuration a second time ------------------------ Aug 17 13:41:26 <user> : LOG5[main]: Reading configuration from file /home/pi/stunnel.conf My configuration file: [PSK client] client = yes accept = <listen address> connect = <remote address> PSKsecrets = /home/pi/psk.txt [PSK server] accept = <listen address> connect = <remote address> PSKsecrets = /home/pi/psk.txt Sven Reschke Embedded System SW Developer G-TS/P Volkswagen Infotainment GmbH Universitätsstr. 140 44799 Bochum Mobile +49 1525-887 2762 sven.reschke(a)volkswagen-infotainment.com<mailto:sven.reschke@volkswagen-infotainment.com> http://www.volkswagen-infotainment.com<http://www.volkswagen-infotainment.com/> Sitz der Gesellschaft: Wolfsburg Amtsgericht Braunschweig: HRB 204780 Geschäftsführung: Bernhard Krauße (Sprecher), Tobias Nadjib

2 2