stunnel-users

3306 discussions

Static Version Built?
by Hal Vaughan 15 Mar '06

15 Mar '06

Okay, I think I have a static version of Stunnel now. Thanks to help and answers from Jan and Keven! I did this: ./configure --with-ssl=/usr --enable-static --disable-shared --disable-libwrap I'm not sure about the "--with-ssl" option, but it didn't seem to want to work without it. So I have a few questions: 1) How can I be sure it is static? 2) If it isn't static and I run it on a system that does not have all it needs, will it give me errors immediately, or only if a subroutine needs a library that isn't there. 3) Does the "--disable-libwrap" or anything else remove any features? 4) Does using the "--with-ssl" option make it use the libs on my system and make it non-static? 5) My "best guess" is that to install a static version, I just copy over the files src/stunnel and src/libstunnel.la to /usr/bin and /usr/lib, respectively. Is that right? (And, of course, copy man files to the right place.) Thanks -- I hope I'll be done with all these questions soon. Hal

2 4

Logging change in 4.15
by MeeAGhost 15 Mar '06

15 Mar '06

Running 4.15 the messages below being written to STDOUT whenever I start stunnel but not written to the the file defined by output in the .conf file. In prior versions the messages were written to whatever I had defined output to equal. 2006.03.14 19:40:03 LOG7[11513:1]: Snagged 64 random bytes from /export/home/ghost/.rnd 2006.03.14 19:40:03 LOG7[11513:1]: Wrote 1024 new random bytes to /export/home/ghost/.rnd 2006.03.14 19:40:03 LOG7[11513:1]: RAND_status claims sufficient entropy for the PRNG 2006.03.14 19:40:03 LOG6[11513:1]: PRNG seeded successfully 2006.03.14 19:40:03 LOG7[11513:1]: Certificate: /usr/local/etc/stunnel/server.pem 2006.03.14 19:40:03 LOG7[11513:1]: Key file: /usr/local/etc/stunnel/server.pem 2006.03.14 19:40:03 LOG7[11513:1]: Loaded verify certificates from /usr/local/etc/stunnel/cacert.pem 2006.03.14 19:40:03 LOG7[11513:1]: Verify directory set to /usr/local/etc/stunnel/certificates 2006.03.14 19:40:03 LOG5[11513:1]: Peer certificate location /usr/local/etc/stunnel/certificates I looked in ctx.c and ssl.c at the calls to s_log (in log.c) but wasn't able to determine what would cause some messages to be written to the output file and the messages above to be written to STDOUT. I think it has to do with file_putline not being passed the output file name. Anyone else seeing this behavior? -- Chris

2 1

What is "lwrap"? (compiling problem)
by Hal Vaughan 15 Mar '06

15 Mar '06

I keep getting a message: /usr/bin/ld: cannot find -lwrap when I try make. I had to use: ./configure --with-ssl=/usr and it worked, but I keep getting the above error message. I've Googled for it and it seems to reference an SSL library, but it looks like I've got them all installed (unless it needs Python SSL libraries). Thanks! Hal

1 0

SSL Libraries on Debian
by Hal Vaughan 14 Mar '06

14 Mar '06

I tried compiling Stunnel on Debian and get an error -- I need the ssl libraries. I tried ./configure --with-ssl=/usr/lib/ssl, which is where libssl-dev was installed, and it doesn't help. There is a libssl file in /usr/lib, but --with-ssl=/usr/lib/ssl doesn't seem to work either. What package on Debian provides the needed headers (according to aptitude, libssl-dev, the one I installed, does, but experience is telling me otherwise). (This, btw, is why I am trying to make static binaries -- you can never tell what to expect on Linux systems at this point. Often config/make doesn't work and there are many times when packages don't work out for one reason or another.) Thanks for any help on this. Hal

3 4

stunnel 4.14 died silently under traffic stress test
by Zhuang Yuyao 14 Mar '06

14 Mar '06

Hi, While doing apache benchmark test, stunnel 4.14 died silently. Network topo: client(192.168.20.81 debian running ab2) ---(https)---> gateway(192.168.20.1 running stunnel in server mode) ---(http)---> web server(192.168.10.254) Gateway hardware: 80386 compatible cpu (VIA C3 400Mhz); RAM: 64M stunnel.conf: ############################### ..... compression = zlib client = no ciphers = DES-CBC3-SHA:DES-CBC3-MD5 verify = 1 [test] accept = 443 connect = 192.168.10.254:80 TIMEOUTclose = 0 ############################### Web Server hardware: Dell 2850 with 2G RAM. test_100k.html is a html file whose size is 100k bytes. test_4k.html is a html file whose size is 4k bytes. In client, I am using ab2 as the stress test tool: #ab2 -n 200 -c 100 https://192.168.20.1:443/test_4k.html (means 200 request and 100 concurrency) successfully finished. #ab2 -n 200 -c 100 https://192.168.20.1:443/test_100k.html (means 200 request and 100 concurrency) stunnel died after a few seconds, running log shows: 2006.03.09 15:20:39 LOG5[6962:1024]: stunnel 4.14 on i386-pc-linux-gnu PTHREAD+POLL+IPv4 with OpenSSL 0.9.7i 14 Oct 2005 2006.03.09 15:20:39 LOG5[6962:1024]: 500 clients allowed 2006.03.09 15:20:53 LOG5[6964:1026]: test connected from 192.168.20.81:50186 2006.03.09 15:20:53 LOG5[6965:2051]: test connected from 192.168.20.81:50187 2006.03.09 15:20:53 LOG5[6966:3076]: test connected from 192.168.20.81:50188 2006.03.09 15:20:53 LOG5[6967:4101]: test connected from 192.168.20.81:50189 2006.03.09 15:20:53 LOG5[6968:5126]: test connected from 192.168.20.81:50190 2006.03.09 15:20:54 LOG5[6969:6151]: test connected from 192.168.20.81:50191 2006.03.09 15:20:54 LOG5[6970:7176]: test connected from 192.168.20.81:50192 2006.03.09 15:20:54 LOG5[6971:8201]: test connected from 192.168.20.81:50193 2006.03.09 15:20:54 LOG5[6972:9226]: test connected from 192.168.20.81:50194 2006.03.09 15:20:54 LOG5[6973:10251]: test connected from 192.168.20.81:50195 2006.03.09 15:20:54 LOG5[6974:11276]: test connected from 192.168.20.81:50196 2006.03.09 15:20:54 LOG5[6975:12301]: test connected from 192.168.20.81:50197 2006.03.09 15:20:54 LOG5[6976:13326]: test connected from 192.168.20.81:50198 2006.03.09 15:20:54 LOG5[6977:14351]: test connected from 192.168.20.81:50199 2006.03.09 15:20:54 LOG5[6978:15376]: test connected from 192.168.20.81:50200 2006.03.09 15:20:54 LOG5[6979:16401]: test connected from 192.168.20.81:50201 2006.03.09 15:20:54 LOG5[6980:17426]: test connected from 192.168.20.81:50202 2006.03.09 15:20:54 LOG5[6981:18451]: test connected from 192.168.20.81:50203 2006.03.09 15:20:54 LOG5[6982:19476]: test connected from 192.168.20.81:50204 2006.03.09 15:20:54 LOG5[6983:20501]: test connected from 192.168.20.81:50205 2006.03.09 15:20:54 LOG5[6984:21526]: test connected from 192.168.20.81:50206 2006.03.09 15:20:54 LOG5[6985:22551]: test connected from 192.168.20.81:50207 2006.03.09 15:20:54 LOG5[6986:23576]: test connected from 192.168.20.81:50208 2006.03.09 15:20:54 LOG5[6987:24601]: test connected from 192.168.20.81:50209 2006.03.09 15:20:54 LOG5[6988:25626]: test connected from 192.168.20.81:50210 2006.03.09 15:20:54 LOG5[6989:26651]: test connected from 192.168.20.81:50211 2006.03.09 15:20:54 LOG5[6990:27676]: test connected from 192.168.20.81:50212 2006.03.09 15:20:54 LOG5[6991:28701]: test connected from 192.168.20.81:50213 2006.03.09 15:20:54 LOG5[6992:29726]: test connected from 192.168.20.81:50214 2006.03.09 15:20:54 LOG5[6993:30751]: test connected from 192.168.20.81:50215 2006.03.09 15:20:54 LOG5[6994:31776]: test connected from 192.168.20.81:50216 2006.03.09 15:20:54 LOG5[6995:32801]: test connected from 192.168.20.81:50217 2006.03.09 15:20:54 LOG5[6996:33826]: test connected from 192.168.20.81:50218 2006.03.09 15:20:54 LOG5[6997:34851]: test connected from 192.168.20.81:50219 2006.03.09 15:20:54 LOG5[6998:35876]: test connected from 192.168.20.81:50220 2006.03.09 15:20:54 LOG5[6999:36901]: test connected from 192.168.20.81:50221 2006.03.09 15:20:54 LOG5[7000:37926]: test connected from 192.168.20.81:50222 2006.03.09 15:20:54 LOG5[7001:38951]: test connected from 192.168.20.81:50223 2006.03.09 15:20:54 LOG5[7002:39976]: test connected from 192.168.20.81:50224 2006.03.09 15:20:54 LOG5[7003:41001]: test connected from 192.168.20.81:50225 2006.03.09 15:20:54 LOG5[7004:42026]: test connected from 192.168.20.81:50226 2006.03.09 15:20:55 LOG5[7005:43051]: test connected from 192.168.20.81:50227 2006.03.09 15:20:55 LOG5[7006:44076]: test connected from 192.168.20.81:50228 2006.03.09 15:20:55 LOG5[7007:45101]: test connected from 192.168.20.81:50229 2006.03.09 15:20:55 LOG5[7008:46126]: test connected from 192.168.20.81:50230 2006.03.09 15:20:55 LOG5[7009:47151]: test connected from 192.168.20.81:50231 2006.03.09 15:20:55 LOG5[7010:48176]: test connected from 192.168.20.81:50232 2006.03.09 15:20:55 LOG5[7011:49201]: test connected from 192.168.20.81:50233 2006.03.09 15:20:55 LOG5[7012:50226]: test connected from 192.168.20.81:50234 2006.03.09 15:20:55 LOG5[7013:51251]: test connected from 192.168.20.81:50235 2006.03.09 15:20:55 LOG5[7014:52276]: test connected from 192.168.20.81:50236 2006.03.09 15:20:55 LOG5[7015:53301]: test connected from 192.168.20.81:50237 2006.03.09 15:20:55 LOG5[7016:54326]: test connected from 192.168.20.81:50238 2006.03.09 15:20:55 LOG5[7017:55351]: test connected from 192.168.20.81:50239 2006.03.09 15:20:55 LOG5[7018:56376]: test connected from 192.168.20.81:50240 2006.03.09 15:20:55 LOG5[7019:57401]: test connected from 192.168.20.81:50241 2006.03.09 15:20:55 LOG5[7020:58426]: test connected from 192.168.20.81:50242 2006.03.09 15:20:55 LOG5[7021:59451]: test connected from 192.168.20.81:50243 2006.03.09 15:20:55 LOG5[7022:60476]: test connected from 192.168.20.81:50244 2006.03.09 15:20:55 LOG5[7023:61501]: test connected from 192.168.20.81:50245 2006.03.09 15:20:55 LOG5[7024:62526]: test connected from 192.168.20.81:50246 Killed It looks like that is resource on gateway is exhausted under such a traffic, but my question is: why stunnel just died instead of rejecting further connections. Thanks very much. Zhuang Yuyao

2 7

Static Binaries for Linux? (Didn't get answer previously)
by Hal Vaughan 13 Mar '06

13 Mar '06

I've brought this up before and didn't get ANY answer at all, and I'm sure someone involved with Stunnel would at least know if what I'm asking is possible. I know help is given on a volunteer basis, so if someone could help me figure this out, it would be a huge help. So far I've spent over a week on getting Stunnel or another program to do the job and if I can't get this to work, I need to just drop Stunnel and figure out how to do the same thing in Java. So thanks for any help. I may need to include Stunnel with my application so my clients behind restrictive firewalls can still enable me to work on their systems with RealVNC. I'm working with clients on both Windows and Linux. Is there a way to get static binaries for Linux? I don't know C or C++ or much more than running ./configure && make && make install. I'd like to know if it is possible or what is needed to create static binaries of the files so I could easily install stunnel on any Linux system by just copying over the proper files. Also, what archive format does the Windows installer use? I'd like to do something similar for Windows, so if I could pull the files out of the installer and copy them over myself, it would be a huge help. Thanks! Hal

3 4

Compile failure with 4.15
by MeeAGhost 13 Mar '06

13 Mar '06

I ran ./configure --enable-dh --disable-libwrap then when I try to execute make the build fails: [...] if gcc -DPACKAGE_NAME=\"stunnel\" -DPACKAGE_TARNAME=\"stunnel\" -DPACKAGE_VERSION=\"4.15\" -DPACKAGE_STRING=\"stunnel\ 4.15\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"stunnel\" -DVERSION=\"4.15\" -DHOST=\"sparc-sun-solaris2.9\" -DCPU_SPARC=1 -DVENDOR_SUN=1 -DOS_SOLARIS2_9=1 -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DSIZEOF_UNSIGNED_CHAR=1 -DSIZEOF_UNSIGNED_SHORT=2 -DSIZEOF_UNSIGNED_INT=4 -DSIZEOF_UNSIGNED_LONG=4 -DHAVE_DEV_PTMX=1 -DHAVE_LIBNSL=1 -DHAVE_LIBSOCKET=1 -DHAVE_LIBDL=1 -DHAVE_LIBZ=1 -DHAVE_UCONTEXT_H=1 -DHAVE_PTHREAD_H=1 -DHAVE_SYS_SELECT_H=1 -DHAVE_POLL_H=1 -DHAVE_SYS_POLL_H=1 -DHAVE_SYS_IOCTL_H=1 -DHAVE_SYS_FILIO_H=1 -DHAVE_STROPTS_H=1 -DHAVE_GRP_H=1 -DHAVE_UNISTD_H=1 -DHAVE_SYS_RESOURCE_H=1 -DHAVE_SNPRINTF=1 -DHAVE_VSNPRINTF=1 -DHAVE_WAITPID=1 -DHAVE_WAIT4=1 -DHAVE_SETSID=1 -DHAVE_SETGROUPS=1 -DHAVE_CHROOT=1 -DHAVE_SYSCONF=1 -DHAVE_GETRLIMIT=1 -DHAVE_PTHREAD_SIGMASK=1 -DHAVE_LOCALTIME_R=1 -DHAVE_GETCONTEXT=1 -DHAVE___MAKECONTEXT_V2=1 -DHAVE_POLL=1 -DHAVE_ENDHOSTENT=1 -DHAVE_GETADDRINFO=1 -DHAVE_GETNAMEINFO=1 -DHAVE_OPENSSL=1 -Dssldir=\"/usr/local\" -DHAVE_OSSL_ENGINE_H=1 -DRANDOM_FILE=\"/dev/urandom\" -DHAVE_LIBPTHREAD=1 -DUSE_PTHREAD=1 -DUSE_DH=1 -I. -I. -I/usr/kerberos/include -DLIBDIR='"/usr/local/lib"' -DCONFDIR='"/usr/local/etc/stunnel"' -DPIDFILE='"/usr/local/var/run/stunnel/stunnel.pid"' -g -O2 -Wall -Wshadow -Wcast-align -Wpointer-arith -I/usr/local/include -MT ctx.o -MD -MP -MF ".deps/ctx.Tpo" -c -o ctx.o ctx.c; \ then mv -f ".deps/ctx.Tpo" ".deps/ctx.Po"; else rm -f ".deps/ctx.Tpo"; exit 1; fi ctx.c: In function `init_dh': ctx.c:170: error: `section' undeclared (first use in this function) ctx.c:170: error: (Each undeclared identifier is reported only once ctx.c:170: error: for each function it appears in.) ctx.c:198: error: `ctx' undeclared (first use in this function) ctx.c: In function `crl_callback': ctx.c:491: warning: cast increases required alignment of target type make[1]: *** [ctx.o] Error 1 make[1]: Leaving directory `/tmp/stunnel-4.15/src' make: *** [all-recursive] Error 1 I checked ctx.c and couldn't see where section and ctx are being defined inside of the init_dh function nor could I see them defined in the header files. Am I missing something? TIA for any assistance. -- Chris

2 1

Rainbow Technologies CryptoSwift CS200 driver
by Michal Trojnara 11 Mar '06

11 Mar '06

Dear Users, I just bought a second-hand Rainbow Technologies CryptoSwift CS200 card to test engine support for stunnel. Unfortunately I can't find any Linux/Windows driver for this card. Do you have one? Please help. Best regards, Mike

2 1

stunnel 4.15 released
by Michal Trojnara 11 Mar '06

11 Mar '06

Version 4.15, 2006.03.11, urgency: LOW: * Release notes - There are a lot of new features in this version. I recommend to test it well before upgrading your mission-critical systems. * Bugfixes - Fix for pthreads on Solaris 10 (thx to Hans Werner Strube <strube(a)physik3.gwdg.de>). - Attempt to autodetect socklen_t type in configure script. - Default threading model changed to pthread for better portability. - DH parameters are not included in the certificate by default. * New features sponsored by Software House http://www.swhouse.com/ - Most SSL-related options (including client, cert, key) are now available on service level, so it is possible to have an SSL client and an SSL server in a single stunnel process. - Windows CE (version 3.0 and higher) support. * New features - Client mode CONNECT protocol support (RFC 2817 section 5.2). http://www.ietf.org/rfc/rfc2817.txt - Retrying exec+connect services added. * File locations are more compliant to Filesystem Hierarchy Standard 2.3 - configuration and certificates are in $prefix/etc/stunnel/ - binaries are in $prefix/sbin/ - default pid file is $prefix/var/run/stunnel.pid - manual is $prefix/man/man8/stunnel.8 - other docs are in $prefix/share/doc/stunnel/ - libstunnel is in $prefix/lib - chroot directory is setup in $prefix/var/lib/stunnel/ this directory is chmoded 1770 and group nogroup sha1sum for stunnel-4.15.tar.gz: 735406c1ca94904581158a434214e1f6568539d0 Home page: http://stunnel.mirt.net/ Download: ftp://stunnel.mirt.net/stunnel/ Best regards, Mike

1 0

"STACK_RESERVE is too high" after every connect
by Rob Munsch 10 Mar '06

10 Mar '06

Hello. I've recently gone on a self-inflicted crash course on openssl and seem to have gotten a centralized logging system going, logging to MySQL. I'm seeing everything i'd expect to see, but with an extra bit: what is STACK_RESERVE, and why is it too high? Some context, mildly sanitized: |Mar 9 18:33:34 isis stunnel: LOG5[24810:6]: syslog-ng connected from x.x.x.x:44032| |Mar 9 18:33:34 isis stunnel: LOG5[24810:6]: VERIFY OK: depth=1, /C=US/ST=Pennsylvania/L=Philadelphia/O=Solutions for Progress/CN=blah.com/emailAddress=blah@blah| |Mar 9 18:33:34 isis stunnel: LOG5[24810:6]: VERIFY OK: depth=0, /C=US/ST=Pennsylvania/O=Solutions for Progress/OU=IT/CN=IT/emailAddress=blah@blah| |Mar 9 20:39:02 isis stunnel: LOG5[24810:6]: Connection closed: 0 bytes sent to SSL, 2891 bytes sent to socket| |Mar 9 20:39:02 isis stunnel: LOG3[24810:6]: STACK_RESERVE is too high| this is daemon.log on the server machine (isis). Messages from the remote host are going to MySQL successfully. But at the end of every apparently-successful transaction, there's this stack complaint. Can't find anything in the archives or via Google for "stunnel stack_reserve," which i found really odd, which is why i'm here... Version output: stunnel 4.14 on i686-pc-linux-gnu UCONTEXT+POLL+IPv4 with OpenSSL 0.9.7g 11 Apr 2005 Thanks! -- Rob Munsch Solutions For Progress IT

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

stunnel-users