- stunnel-users - stunnel.org

Unable to authenticate with Outlook.com servers (SMTP)
by Aerowinder 15 Oct '15

15 Oct '15

Greetings, Running stunnel 5.25 latest beta (2015-10-15), but also had the same issues on 5.24 final. My config file looks like this: debug = debug output = xxxxx client = yes cert = stunnel.pem [outlook-smtp] accept = 127.0.0.1:25 connect = smtp-mail.outlook.com:587 protocol = smtp ;protocolUsername = xxxx ;protocolPassword = xxxx CAfile = ca-certs.pem checkHost = smtp-mail.outlook.com OCSPaia = yes verify = 2 I am testing functionality to make sure that I am able to send mail with my Outlook.com address. I currently have User/Pass field commented out, but I've tried with them not commented out, and the actual values instead of "xxxx" (the app I'm using has fields for username and password). My log: 2015.10.15 11:01:24 LOG7[cron]: Cron started 2015.10.15 11:01:24 LOG7[main]: No limit detected for the number of clients 2015.10.15 11:01:24 LOG5[main]: stunnel 5.25 on x86-pc-msvc-1500 platform 2015.10.15 11:01:24 LOG5[main]: Compiled/running with OpenSSL 1.0.2d-fips 9 Jul 2015 2015.10.15 11:01:24 LOG5[main]: Threading:WIN32 Sockets:SELECT,IPv6 TLS:ENGINE,FIPS,OCSP,PSK,SNI 2015.10.15 11:01:24 LOG7[main]: errno: (*_errno()) 2015.10.15 11:01:24 LOG7[ui]: GUI message loop initialized 2015.10.15 11:01:24 LOG5[main]: Reading configuration from file stunnel.conf 2015.10.15 11:01:24 LOG5[main]: UTF-8 byte order mark detected 2015.10.15 11:01:24 LOG5[main]: FIPS mode disabled 2015.10.15 11:01:24 LOG7[main]: Compression disabled 2015.10.15 11:01:24 LOG7[main]: PRNG seeded successfully 2015.10.15 11:01:24 LOG6[main]: Initializing service [outlook-smtp] 2015.10.15 11:01:24 LOG6[main]: Loading certificate from file: stunnel.pem 2015.10.15 11:01:24 LOG6[main]: Loading key from file: stunnel.pem 2015.10.15 11:01:24 LOG7[main]: Private key check succeeded 2015.10.15 11:01:24 LOG7[main]: SSL options: 0x03000004 (+0x03000000, -0x00000000) 2015.10.15 11:01:24 LOG5[main]: Configuration successful 2015.10.15 11:01:24 LOG7[main]: Listening file descriptor created (FD=652) 2015.10.15 11:01:24 LOG7[main]: Service [outlook-smtp] (FD=652) bound to 127.0.0.1:25 2015.10.15 11:01:31 LOG7[main]: Found 1 ready file descriptor(s) 2015.10.15 11:01:31 LOG7[main]: FD=408 ifds=r-x ofds=--- 2015.10.15 11:01:31 LOG7[main]: Service [outlook-smtp] accepted (FD=692) from 127.0.0.1:58065 2015.10.15 11:01:31 LOG7[main]: Creating a new thread 2015.10.15 11:01:31 LOG7[main]: New thread created 2015.10.15 11:01:31 LOG7[0]: Service [outlook-smtp] started 2015.10.15 11:01:31 LOG5[0]: Service [outlook-smtp] accepted connection from 127.0.0.1:58065 2015.10.15 11:01:31 LOG6[0]: s_connect: connecting 65.55.176.126:587 2015.10.15 11:01:31 LOG7[0]: s_connect: s_poll_wait 65.55.176.126:587: waiting 10 seconds 2015.10.15 11:01:31 LOG5[0]: s_connect: connected 65.55.176.126:587 2015.10.15 11:01:31 LOG5[0]: Service [outlook-smtp] connected remote server from 172.26.1.51:58066 2015.10.15 11:01:31 LOG7[0]: Remote descriptor (FD=704) initialized 2015.10.15 11:01:31 LOG7[0]: <- 220 BLU436-SMTP245.smtp.hotmail.com Microsoft ESMTP MAIL Service, Version: 8.0.9200.16384 ready at Thu, 15 Oct 2015 08:01:34 -0700 2015.10.15 11:01:31 LOG7[0]: -> 220 BLU436-SMTP245.smtp.hotmail.com Microsoft ESMTP MAIL Service, Version: 8.0.9200.16384 ready at Thu, 15 Oct 2015 08:01:34 -0700 2015.10.15 11:01:31 LOG7[0]: -> EHLO localhost 2015.10.15 11:01:31 LOG7[0]: <- 250-BLU436-SMTP245.smtp.hotmail.com Hello [70.90.151.129] 2015.10.15 11:01:31 LOG7[0]: <- 250-TURN 2015.10.15 11:01:31 LOG7[0]: <- 250-SIZE 41943040 2015.10.15 11:01:31 LOG7[0]: <- 250-ETRN 2015.10.15 11:01:31 LOG7[0]: <- 250-PIPELINING 2015.10.15 11:01:31 LOG7[0]: <- 250-DSN 2015.10.15 11:01:31 LOG7[0]: <- 250-ENHANCEDSTATUSCODES 2015.10.15 11:01:31 LOG7[0]: <- 250-8bitmime 2015.10.15 11:01:31 LOG7[0]: <- 250-BINARYMIME 2015.10.15 11:01:31 LOG7[0]: <- 250-CHUNKING 2015.10.15 11:01:31 LOG7[0]: <- 250-VRFY 2015.10.15 11:01:31 LOG7[0]: <- 250-TLS 2015.10.15 11:01:31 LOG7[0]: <- 250-STARTTLS 2015.10.15 11:01:31 LOG7[0]: <- 250 OK 2015.10.15 11:01:31 LOG7[0]: -> STARTTLS 2015.10.15 11:01:31 LOG7[0]: <- 220 2.0.0 SMTP server ready 2015.10.15 11:01:31 LOG6[0]: SNI: sending servername: smtp-mail.outlook.com 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): before/connect initialization 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv2/v3 write client hello A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 read server hello A 2015.10.15 11:01:31 LOG7[0]: Verification started at depth=2: C=BE, O=GlobalSign nv-sa, OU=Root CA, CN=GlobalSign Root CA 2015.10.15 11:01:31 LOG7[0]: CERT: Pre-verification succeeded 2015.10.15 11:01:31 LOG7[0]: OCSP: Ignoring root certificate 2015.10.15 11:01:31 LOG6[0]: Certificate accepted at depth=2: C=BE, O=GlobalSign nv-sa, OU=Root CA, CN=GlobalSign Root CA 2015.10.15 11:01:31 LOG7[0]: Verification started at depth=1: C=BE, O=GlobalSign nv-sa, CN=GlobalSign Organization Validation CA - G2 2015.10.15 11:01:31 LOG7[0]: CERT: Pre-verification succeeded 2015.10.15 11:01:31 LOG5[0]: OCSP: Connecting the AIA responder "http://ocsp.globalsign.com/rootr1" 2015.10.15 11:01:31 LOG6[0]: s_connect: connecting 108.162.232.204:80 2015.10.15 11:01:31 LOG7[0]: s_connect: s_poll_wait 108.162.232.204:80: waiting 10 seconds 2015.10.15 11:01:31 LOG5[0]: s_connect: connected 108.162.232.204:80 2015.10.15 11:01:31 LOG7[0]: OCSP: Connected ocsp.globalsign.com:80 2015.10.15 11:01:31 LOG7[0]: OCSP: Response received 2015.10.15 11:01:31 LOG6[0]: OCSP: Status: good 2015.10.15 11:01:31 LOG6[0]: OCSP: This update: Oct 15 10:27:35 2015 GMT 2015.10.15 11:01:31 LOG6[0]: OCSP: Next update: Oct 19 10:27:35 2015 GMT 2015.10.15 11:01:31 LOG5[0]: OCSP: Certificate accepted 2015.10.15 11:01:31 LOG6[0]: Certificate accepted at depth=1: C=BE, O=GlobalSign nv-sa, CN=GlobalSign Organization Validation CA - G2 2015.10.15 11:01:31 LOG7[0]: Verification started at depth=0: C=US, ST=Washington, L=Redmond, O=Microsoft Corporation, CN=*.hotmail.com 2015.10.15 11:01:31 LOG7[0]: CERT: Pre-verification succeeded 2015.10.15 11:01:31 LOG6[0]: CERT: Host name "smtp-mail.outlook.com" matched with "*.outlook.com" 2015.10.15 11:01:31 LOG5[0]: OCSP: Connecting the AIA responder "http://ocsp2.globalsign.com/gsorganizationvalg2" 2015.10.15 11:01:31 LOG6[0]: s_connect: connecting 108.162.232.196:80 2015.10.15 11:01:31 LOG7[0]: s_connect: s_poll_wait 108.162.232.196:80: waiting 10 seconds 2015.10.15 11:01:31 LOG5[0]: s_connect: connected 108.162.232.196:80 2015.10.15 11:01:31 LOG7[0]: OCSP: Connected ocsp2.globalsign.com:80 2015.10.15 11:01:31 LOG7[0]: OCSP: Response received 2015.10.15 11:01:31 LOG6[0]: OCSP: Status: good 2015.10.15 11:01:31 LOG6[0]: OCSP: This update: Oct 15 08:26:15 2015 GMT 2015.10.15 11:01:31 LOG6[0]: OCSP: Next update: Oct 16 08:26:15 2015 GMT 2015.10.15 11:01:31 LOG5[0]: OCSP: Certificate accepted 2015.10.15 11:01:31 LOG5[0]: Certificate accepted at depth=0: C=US, ST=Washington, L=Redmond, O=Microsoft Corporation, CN=*.hotmail.com 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 read server certificate A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 read server key exchange A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 read server done A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 write client key exchange A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 write change cipher spec A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 write finished A 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 flush data 2015.10.15 11:01:31 LOG7[0]: SSL state (connect): SSLv3 read finished A 2015.10.15 11:01:31 LOG7[0]: 1 client connect(s) requested 2015.10.15 11:01:31 LOG7[0]: 1 client connect(s) succeeded 2015.10.15 11:01:31 LOG7[0]: 0 client renegotiation(s) requested 2015.10.15 11:01:31 LOG7[0]: 0 session reuse(s) 2015.10.15 11:01:31 LOG6[0]: SSL connected: new session negotiated 2015.10.15 11:01:31 LOG7[0]: Peer certificate was cached (3461 bytes) 2015.10.15 11:01:31 LOG6[0]: Negotiated TLSv1.2 ciphersuite ECDHE-RSA-AES256-SHA384 (256-bit encryption) 2015.10.15 11:01:31 LOG7[0]: Compression: null, expansion: null 2015.10.15 11:01:31 LOG6[0]: SSL socket closed (SSL_read) 2015.10.15 11:01:31 LOG7[0]: Sent socket write shutdown 2015.10.15 11:01:31 LOG5[0]: Connection closed: 139 byte(s) sent to SSL, 351 byte(s) sent to socket 2015.10.15 11:01:31 LOG7[0]: Remote descriptor (FD=704) closed 2015.10.15 11:01:31 LOG7[0]: Local descriptor (FD=692) closed 2015.10.15 11:01:31 LOG7[0]: Service [outlook-smtp] finished (0 left) I don't see any errors in the log, but Outlook is reporting an incorrect password on the security section of my Microsoft Account. More specifically, I'm seeing error 5.7.3: Requested action aborted, user not authenticated," I know the password works because I set it up as an account in Outlook 2010 and tested sending the email. It works great. I am using 2FA with an app-specific password. But I have also tried without 2FA. I am 100% certain the account name xxxx(a)outlook.com and the password are correct. As far as the config file goes, I've tried just about everything I could find on Google. Nothing works, always the same error message. I also experienced similar failures with Gmail. Any ideas?

2 1

prng_init failure
by Philip Craig 15 Oct '15

15 Oct '15

prng_init() is failing in 5.24 when cross-compiling with openssl due to a combination of two reasons: 1. The --with-random configure option is not supported when cross compiling. 2. The hardcoded /dev/urandom is no longer used if OPENSSL_NO_EGD is not defined. The following patch reverts the behaviour to how it was in 5.23, but probably the --with-random configure option should be supported too. --- stunnel-5.25.orig/src/ssl.c +++ stunnel-5.25/src/ssl.c @@ -207,7 +207,8 @@ NOEXPORT int prng_init(GLOBAL_OPTIONS *g return 0; /* success */ } s_log(LOG_DEBUG, "RAND_screen failed to sufficiently seed PRNG"); -#elif !defined(OPENSSL_NO_EGD) +#else +#ifndef OPENSSL_NO_EGD if(global->egd_sock) { if((bytes=RAND_egd(global->egd_sock))==-1) { s_log(LOG_WARNING, "EGD Socket %s failed", global->egd_sock); @@ -220,7 +221,7 @@ NOEXPORT int prng_init(GLOBAL_OPTIONS *g so no need to check if seeded sufficiently */ } } -#else +#endif /* try the good-old default /dev/urandom, if available */ totbytes+=add_rand_file(global, "/dev/urandom"); if(RAND_status())

2 1

[PATCH] only use aysnc-signal-safe functions after fork
by Philip Craig 15 Oct '15

15 Oct '15

POSIX requires that multi-threaded processes only use async-signal-safe functions after calling fork(). For more info on the problems this avoids, see the rationale at: http://pubs.opengroup.org/onlinepubs/9699919799/functions/pthread_atfork.ht… The specific problem this patch solves is that, when there are multiple simultaneous requests, stunnel is often hanging on a futex syscall when a child process calls free as a result of tls_alloc(). This problem is experienced when using uClibc. I haven't tested, but I believe the malloc implementation in glibc does not have this problem. --- stunnel-5.24.orig/src/client.c +++ stunnel-5.24/src/client.c @@ -1100,6 +1100,8 @@ NOEXPORT SOCKET connect_local(CLI *c) { NOEXPORT SOCKET connect_local(CLI *c) { /* spawn local process */ char *name, host[40], port[6]; + char *env[10]; + int i, envc; int fd[2], pid; X509 *peer_cert; #ifdef HAVE_PTHREAD_SIGMASK @@ -1115,19 +1117,55 @@ NOEXPORT SOCKET connect_local(CLI *c) { } else if(make_sockets(fd)) longjmp(c->err, 1); + set_nonblock(fd[1], 0); /* switch back to blocking mode */ + + envc = 0; + + if(!getnameinfo(&c->peer_addr.sa, c->peer_addr_len, + host, 40, port, 6, NI_NUMERICHOST|NI_NUMERICSERV)) { + /* just don't set these variables if getnameinfo() fails */ + env[envc++] = str_printf("REMOTE_HOST=%s", host); + env[envc++] = str_printf("REMOTE_PORT=%s", port); + if(c->opt->option.transparent_src) { +#ifndef LIBDIR +#define LIBDIR "." +#endif +#ifdef MACH64 + env[envc++] = str_dup("LD_PRELOAD_32=" LIBDIR "/libstunnel.so"); + env[envc++] = str_dup("LD_PRELOAD_64=" LIBDIR "/" MACH64 "/libstunnel.so"); +#elif __osf /* for Tru64 _RLD_LIST is used instead */ + env[envc++] = str_dup("_RLD_LIST=" LIBDIR "/libstunnel.so:DEFAULT"); +#else + env[envc++] = str_dup("LD_PRELOAD=" LIBDIR "/libstunnel.so"); +#endif + } + } + + if(c->ssl) { + peer_cert=SSL_get_peer_certificate(c->ssl); + if(peer_cert) { + name=X509_NAME2text(X509_get_subject_name(peer_cert)); + env[envc++] = str_printf("SSL_CLIENT_DN=%s", name); + name=X509_NAME2text(X509_get_issuer_name(peer_cert)); + env[envc++] = str_printf("SSL_CLIENT_I_DN=%s", name); + X509_free(peer_cert); + } + } + + env[envc] = NULL; pid=fork(); c->pid=(unsigned long)pid; switch(pid) { case -1: /* error */ + for (i = 0; i < envc; i++) + str_free(env[i]); closesocket(fd[0]); closesocket(fd[1]); ioerror("fork"); longjmp(c->err, 1); case 0: /* child */ - tls_alloc(NULL, c->tls, NULL); /* reuse thread-local storage */ closesocket(fd[0]); - set_nonblock(fd[1], 0); /* switch back to blocking mode */ /* dup2() does not copy FD_CLOEXEC flag */ dup2(fd[1], 0); dup2(fd[1], 1); @@ -1135,36 +1173,6 @@ NOEXPORT SOCKET connect_local(CLI *c) { dup2(fd[1], 2); closesocket(fd[1]); /* not really needed due to FD_CLOEXEC */ - if(!getnameinfo(&c->peer_addr.sa, c->peer_addr_len, - host, 40, port, 6, NI_NUMERICHOST|NI_NUMERICSERV)) { - /* just don't set these variables if getnameinfo() fails */ - putenv(str_printf("REMOTE_HOST=%s", host)); - putenv(str_printf("REMOTE_PORT=%s", port)); - if(c->opt->option.transparent_src) { -#ifndef LIBDIR -#define LIBDIR "." -#endif -#ifdef MACH64 - putenv("LD_PRELOAD_32=" LIBDIR "/libstunnel.so"); - putenv("LD_PRELOAD_64=" LIBDIR "/" MACH64 "/libstunnel.so"); -#elif __osf /* for Tru64 _RLD_LIST is used instead */ - putenv("_RLD_LIST=" LIBDIR "/libstunnel.so:DEFAULT"); -#else - putenv("LD_PRELOAD=" LIBDIR "/libstunnel.so"); -#endif - } - } - - if(c->ssl) { - peer_cert=SSL_get_peer_certificate(c->ssl); - if(peer_cert) { - name=X509_NAME2text(X509_get_subject_name(peer_cert)); - putenv(str_printf("SSL_CLIENT_DN=%s", name)); - name=X509_NAME2text(X509_get_issuer_name(peer_cert)); - putenv(str_printf("SSL_CLIENT_I_DN=%s", name)); - X509_free(peer_cert); - } - } #ifdef HAVE_PTHREAD_SIGMASK sigemptyset(&newmask); sigprocmask(SIG_SETMASK, &newmask, NULL); @@ -1176,11 +1184,13 @@ NOEXPORT SOCKET connect_local(CLI *c) { signal(SIGTERM, SIG_DFL); signal(SIGQUIT, SIG_DFL); signal(SIGINT, SIG_DFL); - execvp(c->opt->exec_name, c->opt->exec_args); + execve(c->opt->exec_name, c->opt->exec_args, env); ioerror(c->opt->exec_name); /* execvp failed */ _exit(1); default: /* parent */ s_log(LOG_INFO, "Local mode child started (PID=%lu)", c->pid); + for (i = 0; i < envc; i++) + str_free(env[i]); closesocket(fd[1]); return fd[0]; }

2 2

Re: [stunnel-users] (no subject)
by Adrián Mihálko 10 Oct '15

10 Oct '15

Some good news, I remove client = yes as you suggested: 2015.10.09 12:39:29 LOG5[main]: Configuration successful 2015.10.09 12:39:29 LOG5[main]: Logging to C:\Users\adrianmihalko\AppData\Local\stunnel.log 2015.10.09 12:39:34 LOG6[57]: SSL socket closed (SSL_read) 2015.10.09 12:39:34 LOG5[57]: Connection closed: 0 byte(s) sent to SSL, 445 byte(s) sent to socket 2015.10.09 12:39:34 LOG5[60]: Service [myservice] accepted connection from 192.168.1.25:49671 2015.10.09 12:39:34 LOG6[60]: SSL accepted: new session negotiated 2015.10.09 12:39:34 LOG6[60]: No peer certificate received 2015.10.09 12:39:34 LOG6[60]: Negotiated TLSv1.2 ciphersuite ECDHE-RSA-AES128-GCM-SHA256 (128-bit encryption) 2015.10.09 12:39:34 LOG6[60]: failover: round-robin, starting at entry #0 2015.10.09 12:39:34 LOG6[60]: s_connect: connecting ::1:41952 2015.10.09 12:39:34 LOG5[60]: s_connect: connected ::1:41952 2015.10.09 12:39:34 LOG6[60]: persistence: ::1:41952 cached 2015.10.09 12:39:34 LOG5[60]: Service [myservice] connected remote server from ::1:50598 2015.10.09 12:39:34 LOG6[60]: SSL socket closed (SSL_read) 2015.10.09 12:39:34 LOG5[60]: Connection closed: 0 byte(s) sent to SSL, 0 byte(s) sent to socket 2015.10.09 12:39:34 LOG5[61]: Service [myservice] accepted connection from 192.168.1.25:49672 2015.10.09 12:39:34 LOG6[61]: SSL accepted: new session negotiated 2015.10.09 12:39:34 LOG6[61]: No peer certificate received 2015.10.09 12:39:34 LOG6[61]: Negotiated TLSv1.2 ciphersuite ECDHE-RSA-AES128-GCM-SHA256 (128-bit encryption) 2015.10.09 12:39:34 LOG6[61]: failover: round-robin, starting at entry #1 2015.10.09 12:39:34 LOG6[61]: s_connect: connecting 127.0.0.1:41952 2015.10.09 12:39:34 LOG5[61]: s_connect: connected 127.0.0.1:41952 2015.10.09 12:39:34 LOG6[61]: persistence: 127.0.0.1:41952 cached 2015.10.09 12:39:34 LOG5[61]: Service [myservice] connected remote server from 127.0.0.1:50599 openssl_client log: http://pastebin.com/7bg3sf7J The problem is now that the site loads forever, nothing happens. (this certificate (:1988) is other than the original (:41952). This is not problem? curl test: $ curl https://192.168.1.17:1988/DYMO/DLS/Printing/Check -vk * Trying 192.168.1.17... * Connected to 192.168.1.17 (192.168.1.17) port 1988 (#0) * TLS 1.2 connection using TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 * Server certificate: localhost > GET /DYMO/DLS/Printing/Check HTTP/1.1 > Host: 192.168.1.17:1988 > User-Agent: curl/7.43.0 > Accept: */* > waiting forever. 2015-10-09 12:34 GMT+02:00 Adrián Mihálko <adriankoooo(a)gmail.com>: > In the first mail I wrote ports bad, of course in the log I am using the > good ones. > > [myservice] > cert = stunnel.pem > client = yes > accept = 0.0.0.0:1988 > connect = localhost:41952 > > > 2015-10-09 12:32 GMT+02:00 Adrián Mihálko <adriankoooo(a)gmail.com>: > >> Sorry, curl was only for testing. >> >> Adrians-MacBook-Pro:~ adrianmihalko$ openssl s_client -connect >> 192.168.1.17:1988 >> CONNECTED(00000003) >> 1130:error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown >> protocol:/BuildRoot/Library/Caches/com.apple.xbs/Sources/OpenSSL098/OpenSSL098-59/src/ssl/s23_clnt.c:618: >> >> 2015.10.09 12:23:21 LOG5[main]: Reading configuration from file >> stunnel.conf >> 2015.10.09 12:23:21 LOG5[main]: UTF-8 byte order mark detected >> 2015.10.09 12:23:21 LOG5[main]: FIPS mode disabled >> 2015.10.09 12:23:21 LOG6[main]: Initializing service [gmail-pop3] >> 2015.10.09 12:23:21 LOG6[main]: Initializing service [gmail-imap] >> 2015.10.09 12:23:21 LOG6[main]: Initializing service [gmail-smtp] >> 2015.10.09 12:23:21 LOG6[main]: Initializing service [myservice] >> 2015.10.09 12:23:21 LOG6[main]: Loading certificate from file: stunnel.pem >> 2015.10.09 12:23:21 LOG6[main]: Loading key from file: stunnel.pem >> 2015.10.09 12:23:21 LOG4[main]: Service [myservice] needs authentication >> to prevent MITM attacks >> 2015.10.09 12:23:21 LOG5[main]: Configuration successful >> 2015.10.09 12:23:21 LOG5[main]: Logging to >> C:\Users\adrianmihalko\AppData\Local\stunnel.log >> 2015.10.09 12:23:42 LOG5[39]: Service [myservice] accepted connection >> from 192.168.1.25:49454 >> 2015.10.09 12:23:42 LOG6[39]: failover: round-robin, starting at entry #0 >> 2015.10.09 12:23:42 LOG6[39]: s_connect: connecting ::1:41952 >> 2015.10.09 12:23:42 LOG5[39]: s_connect: connected ::1:41952 >> 2015.10.09 12:23:42 LOG5[39]: Service [myservice] connected remote server >> from ::1:50564 >> 2015.10.09 12:23:42 LOG6[39]: SNI: sending servername: localhost >> 2015.10.09 12:23:42 LOG6[39]: Certificate verification disabled >> 2015.10.09 12:23:42 LOG6[39]: Certificate verification disabled >> 2015.10.09 12:23:42 LOG6[39]: SSL connected: new session negotiated >> 2015.10.09 12:23:42 LOG6[39]: Negotiated TLSv1 ciphersuite AES128-SHA >> (128-bit encryption) >> 2015.10.09 12:23:42 LOG6[39]: SSL socket closed (SSL_read) >> 2015.10.09 12:23:42 LOG5[39]: Connection closed: 130 byte(s) sent to SSL, >> 505 byte(s) sent to socket >> >> If I am connecting to the :41952: >> >> openssl s_client -connect 192.168.1.17:41952 >> ... >> >> --- >> No client certificate CA names sent >> --- >> SSL handshake has read 1724 bytes and written 712 bytes >> --- >> New, TLSv1/SSLv3, Cipher is AES128-SHA >> Server public key is 4096 bit >> Secure Renegotiation IS supported >> Compression: NONE >> Expansion: NONE >> SSL-Session: >> Protocol : TLSv1 >> Cipher : AES128-SHA >> ... >> >> >> 2015-10-09 10:55 GMT+02:00 test rig <testrig(a)z1p.biz>: >> >>> >>> Ouch #2 missing... >>> >>> Hi Adrian, looks good to me so far - mostly. Try to replace the >>> client=yes with a client=no on the server >>> >>> You are connection to :9999 with curl(?) >>> Try verify it via "openssl s_client -connect yourserverip:1988" command >>> >>> Best Regards >>> Michael >>> >>> --- Ursprüngliche Nachricht --- >>> *Von:* "test rig" <testrig(a)z1p.biz> >>> *Datum:* 09.10.2015 09:48:02 >>> *An:* "stunnel-users(a)stunnel.org." <stunnel-users(a)stunnel.org> >>> *Betreff:* Re: [stunnel-users] (no subject) >>> >>> Hi Adrian, looks good to me so far - mostly. Try to replace the >>> client=yes with a client=no on the server >>> >>> --- Ursprüngliche Nachricht --- >>> *Von:* Adrián Mihálko >>> *Datum:* 09.10.2015 08:15:19 >>> *An:* stunnel-users(a)stunnel.org >>> *Betreff:* [stunnel-users] (no subject) >>> >>> Dear stunnel users, >>> >>> I have a little service which listen only on https://localhost:4952 and >>> checks source hostname. I want to connect on "listen:1988" and redirect >>> requests with stunnel to "localhost:4952" >>> >>> https://192.168.1.10:1988 -> redirect https://localhost:4952 >>> >>> >>> I am trying to configure stunnel like this >>> >>> [myservice] >>> cert = stunnel.pem >>> client = yes >>> accept = 0.0.0.0:1988 >>> connect = localhost:4952 >>> >>> remote machine$ curl https://192.168.1.25:9999/DYMO/DLS/Printing/Check >>> -v >>> * Trying 192.168.1.25... >>> * Connected to 192.168.1.25 (192.168.1.25) port 9999 (#0) >>> * WARNING: using IP address, SNI is being disabled by the OS. >>> * Unknown SSL protocol error in connection to 192.168.1.25:-9847 >>> * Closing connection 0 >>> curl: (35) Unknown SSL protocol error in connection to 192.168.1.25: >>> -9847 >>> >>> stunnel.log: >>> 2015.10.09 09:05:42 LOG5[38]: Service [myservice] accepted connection >>> from 192.168.1.24:60748 >>> 2015.10.09 09:05:42 LOG6[38]: failover: round-robin, starting at entry #1 >>> 2015.10.09 09:05:42 LOG6[38]: s_connect: connecting 127.0.0.1:41952 >>> 2015.10.09 09:05:42 LOG5[38]: s_connect: connected 127.0.0.1:41952 >>> 2015.10.09 09:05:42 LOG5[38]: Service [myservice] connected remote >>> server from 127.0.0.1:50503 >>> 2015.10.09 09:05:42 LOG6[38]: SNI: sending servername: localhost >>> 2015.10.09 09:05:42 LOG6[38]: Certificate verification disabled >>> 2015.10.09 09:05:42 LOG6[38]: Certificate verification disabled >>> 2015.10.09 09:05:42 LOG6[38]: SSL connected: new session negotiated >>> 2015.10.09 09:05:42 LOG6[38]: Negotiated TLSv1 ciphersuite AES128-SHA >>> (128-bit encryption) >>> 2015.10.09 09:05:42 LOG6[38]: SSL socket closed (SSL_read) >>> 2015.10.09 09:05:42 LOG5[38]: Connection closed: 230 byte(s) sent to >>> SSL, 505 byte(s) sent to socket >>> >>> I am tried verify = 1 to 4, either works. :( >>> >>> Best Regards, >>> Adrian >>> >>> >>> >>> ______________________________________________________ >>> powered by Perfect-Privacy.com / Secure-Mail.biz - anonymous and secure >>> internet. >>> >>> >>> >>> ______________________________________________________ >>> powered by Perfect-Privacy.com / Secure-Mail.biz - anonymous and secure >>> internet. >>> >>> _______________________________________________ >>> stunnel-users mailing list >>> stunnel-users(a)stunnel.org >>> https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users >>> >>> >> >

3 6

(no subject)
by Adrián Mihálko 09 Oct '15

09 Oct '15

Dear stunnel users, I have a little service which listen only on https://localhost:4952 and checks source hostname. I want to connect on "listen:1988" and redirect requests with stunnel to "localhost:4952" https://192.168.1.10:1988 -> redirect https://localhost:4952 I am trying to configure stunnel like this [myservice] cert = stunnel.pem client = yes accept = 0.0.0.0:1988 connect = localhost:4952 remote machine$ curl https://192.168.1.25:9999/DYMO/DLS/Printing/Check -v * Trying 192.168.1.25... * Connected to 192.168.1.25 (192.168.1.25) port 9999 (#0) * WARNING: using IP address, SNI is being disabled by the OS. * Unknown SSL protocol error in connection to 192.168.1.25:-9847 * Closing connection 0 curl: (35) Unknown SSL protocol error in connection to 192.168.1.25:-9847 stunnel.log: 2015.10.09 09:05:42 LOG5[38]: Service [myservice] accepted connection from 192.168.1.24:60748 2015.10.09 09:05:42 LOG6[38]: failover: round-robin, starting at entry #1 2015.10.09 09:05:42 LOG6[38]: s_connect: connecting 127.0.0.1:41952 2015.10.09 09:05:42 LOG5[38]: s_connect: connected 127.0.0.1:41952 2015.10.09 09:05:42 LOG5[38]: Service [myservice] connected remote server from 127.0.0.1:50503 2015.10.09 09:05:42 LOG6[38]: SNI: sending servername: localhost 2015.10.09 09:05:42 LOG6[38]: Certificate verification disabled 2015.10.09 09:05:42 LOG6[38]: Certificate verification disabled 2015.10.09 09:05:42 LOG6[38]: SSL connected: new session negotiated 2015.10.09 09:05:42 LOG6[38]: Negotiated TLSv1 ciphersuite AES128-SHA (128-bit encryption) 2015.10.09 09:05:42 LOG6[38]: SSL socket closed (SSL_read) 2015.10.09 09:05:42 LOG5[38]: Connection closed: 230 byte(s) sent to SSL, 505 byte(s) sent to socket I am tried verify = 1 to 4, either works. :( Best Regards, Adrian

2 3

Stunnel for windows dependencies
by test rig 09 Oct '15

09 Oct '15

2 2

stunnel 5.24 released
by Michal Trojnara 08 Oct '15

08 Oct '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Dear Users, I have released version 5.24 of stunnel. The ChangeLog entry: Version 5.24, 2015.10.08, urgency: MEDIUM * New features - Custom CRL verification was replaced with the internal OpenSSL functionality. - *BSD support for "transparent = destination" and client-side "protocol = socks". This feature should work at least on FreeBSD, OpenBSD and OS X. - Added a new "protocolDomain" option for the NTLM authentication (thx to Andreas Botsikas). - Improved compatibility of the NTLM phase 1 message (thx to Andreas Botsikas). - "setuid" and "setgid" options are now also available in service sections. They can be used to set owner and group of the Unix socket specified with "accept". - Added support for the new OpenSSL 1.0.2 SSL options. - Added OPENSSL_NO_EGD support (thx to Bernard Spil). - VC autodetection added to makew32.bat (thx to Andreas Botsikas). * Bugfixes - Fixed the RESOLVE [F0] TOR extension support in SOCKS5. - Fixed the error code reported on the failed bind() requests. - Fixed the sequential log id with the FORK threading. - Restored the missing Microsoft.VC90.CRT.manifest file. Home page: https://www.stunnel.org/ Download: https://www.stunnel.org/downloads.html SHA-256 hashes: ab2e5a1034d422951ddad21b572eb7fa8efb4c4ce04bc86536c6845f3d02b07e stunnel-5.24.tar.gz f6c38d51c2708f3eddbad651091bf0b59e4149a1d0c1e3b227b033f126c6dbee stunnel-5.24-installer.exe ed2a2c5f280970bd5f5efcecbc1dbbd06f66efb68102af31c27d33e3beb48bb5 stunnel-5.24-android.zip Best regards, Mike -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJWFoHsAAoJEC78f/DUFuAUQrAP/iUz2fHvvjkIcJvoR2kMeJfZ NFHE/zGIP6qHbf5Fe5SvOA/ygzQcBppa+FJN9S0uL5z0rLkDghqeWoFzPfLIVrfM y2IJ4SZ1HEESyjaZkCxY0C7Ow1kKdlMPLHv81p+Tk/Zm4qKJ/lIkP+q7XMbNmWjW BOYBLeWPtro1OjKm7v07VkzRGJupj1PHjUgPDAQRxdzqsl6fO7xY9n8b9Nh7Cwc2 1mJhsx8J40M7abftYwpG6Mycu2EOo0lIJ9rm6xZahyekzatsaIhlw2o5gx5GWJvN xj9zpkxntTVQysgP4QVdJZ+Kl17ZZmOS4k+3OqXDgdt51Ck2ty46y9NbhkulIUA8 BzH3KXZCaiBFb/sPdBtW8qmvv0azJLR0b1HfKsZQrtIB/SH0slB9rgsLkidjekKG n7FHf5dlncfAbf8KstJRv41PTYcbtQMGI/+sre2YsqG2/MSJgPfCL2Wj0bTOSmOR 7t6jHX6SItnWtAJ4+3yhlh33B4WYOYBbu3FA3Pw7Wgxur9AOV63MH/Mf96KdMFGS dqoLQh+fp/KoSw1A9OUREtQnxHnbF5MMoU4Z1pTO0FxvdCMnYCg4okiwLqbOKcuM J0T543Deq9XU7v9jDkcgUyQzBYU8lG+G9UdSH11B8phuP7pBUQ35cd62Fl9NWv9I zNjRZIOnbBp6F6hqilb4 =F+az -----END PGP SIGNATURE-----

1 0

Log File Not Appending
by K Monz 06 Oct '15

06 Oct '15

Hi, I am a new STUNNEL user, so I apologize if this is a ridiculously simple and insignificant problem. I have STUNNEL running in a Windows 7 (32 bit) environment, and have the following in stunnel.conf: output = C:\STUNNEL\stunnel.log log = append The log file is placed where I've asked, but the contents represent only the most current STUNNEL execution. I am trying to have a running log in this file. STUNNEL is, of course, doing what it is supposed to do, it is only the logging that has me puzzled. Thanks, in advance.

2 1

Error on Windows 2008r2
by Chris 01 Oct '15

01 Oct '15

Just downloaded and installed the latest windows version. When trying to run stunnel I get the following error: The application has failed to start because its side-by-side configuration is incorrect..... When I check the application event view logs I see "Activation context generation failed for "C:\Program Files (x86)\stunnel\bin\stunnel.exe". Dependent Assembly Microsoft.VC90.CRT,processorArchitecture="x86",publicKeyToken="1fc8b3b9a1e18e3b",type="win32",version="9.0.21022.8" could not be found. Please use sxstrace.exe for detailed diagnosis." It seems I might be missing a pre-req but I can't not determine what that is. Any ideas? Thanks, Chris

1 0

Support for DTLS
by Kanungo, Rajesh 25 Sep '15

25 Sep '15

Hi, I am looking for support for a DTLS proxy. Does Stunnel support DTLS and, if yes, how do I configure it? Here is the scenario: 1. Remote Server supports DTLS. 2. My client does not support DTLS (yet). Support all our platforms is a pretty arduous task. Possible solutions: 1. Ask customer to use Stunnel, drop DTLS and use plain UDP, and use Stunnel/TCP. Other clients support only DTLS so this does not work. 2. Use VPN. Requires customer configuration. Less painful than 1. 3. Use a UDP-DTLS proxy. Regards Rajesh Kanungo North America Security CTO Department, Itron, Inc. (408) 431-3035 Rajesh.kanungo(a)itron.com

1 0

Missing pid from log file since ~5.19
by Philippe Anctil 17 Sep '15

17 Sep '15

Hi, After upgrading to stunnel 5.19 I noticed the log file no long reports the pid of stunnel processes. Instead I find the constant string [0]. Sep 14 12:15:30 hostname stunnel: LOG6[0]: SSL closed (SSL_read) This information is very useful to isolate individual connections. Do later stunnel versions report the pid correctly? I am using fork mode. My syslog is defined as local6.debug and stunnel.conf contains debug=local6.info. Thank you. -- Philippe Anctil

2 3

How to install CA at client side?
by MingHeng Wang 15 Sep '15

15 Sep '15

Hello Stunnel maintainers, I try to use real certificates of my web server for stunnel. I combine private key, my site's cert, and ca-bundle into a pem file, and it works fine when the client doesn't verify any certificate. Then I specify CAfile which is the ca bundle file from my registrar, at client side and turn on verification and always get errors below, whatever level 2 or 3: Sep 15 14:53:28 y400 stunnel[11666]: LOG5[11]: Service [http-proxy3] connected remote server from 192.168.1.104:45746 Sep 15 14:53:28 y400 stunnel[11666]: LOG4[11]: CERT: Pre-verification error: unable to get issuer certificate Sep 15 14:53:28 y400 stunnel[11666]: LOG4[11]: Rejected by CERT at depth=2: However, level 4 works. I want to prevent man-in-middle-attack, so can level 4 achieve that regarding to my current setup? Both server and client side use stunnel 5.17 which are fairly recent.

2 1

Windows 2012 R2 Error on startup
by Han-Earl Guhm 13 Sep '15

13 Sep '15

I installed the latest version of stunnel and am having problems starting it up. I have attached an error log and my config file. I tested on Windows 7 and it worked with no problems, but I would really like it to work on my Windows 2012 server. Thank you, Han-Earl Guhm Coventry Christian Schools hanguhm(a)educatingforlife.com

2 1

The Error with Engine
by Donne, Ann 08 Sep '15

08 Sep '15

Hi all, Need your help. When I configured the Stunnel without the Engine LunaCA3 (product from SafeNet), everything worked fine. After add the global option with the engine: engine=LunaCA3 engineCtrl=SO_PATH:/usr/local/ssl/lib/engines/liblunaca3.so I got the following error. I also searched previous engine issues and saw someone else with the similar issues. Hope that the issue was resolved. The Stunnel and the open SSL info: stunnel 5.20 on x86_64 Linux. Compiled/running with OpenSSL 1.0.1i-fips. The setting is for server with sslVersion = TLSv1 and ciphers = AES128-SHA. 2015.09.08 11:11:01 LOG7[0]: SSL state (accept): SSLv3 read client certificate A 2015.09.08 11:11:01 LOG7[0]: SSL state (accept): SSLv3 read client key exchange A 2015.09.08 11:11:01 LOG7[0]: SSL state (accept): SSLv3 read certificate verify A 2015.09.08 11:11:01 LOG7[0]: SSL alert (write): fatal: bad record mac 2015.09.08 11:11:01 LOG3[0]: SSL_accept: 1408F119: error:1408F119:SSL routines:SSL3_GET_RECORD:decryption failed or bad record mac 2015.09.08 11:11:01 LOG5[0]: Connection reset: 0 byte(s) sent to SSL, 0 byte(s) sent to socket 2015.09.08 11:11:01 LOG7[0]: Deallocating application specific data for addr index 2015.09.08 11:11:01 LOG7[0]: Local socket (FD=3) closed Any help and info are greatly appreciated!! Thanks Ann Donne ---------------------------------------------------------------------- This message, and any attachments, is for the intended recipient(s) only, may contain information that is privileged, confidential and/or proprietary and subject to important terms and conditions available at http://www.bankofamerica.com/emaildisclaimer. If you are not the intended recipient, please delete this message.

1 0

stunnel 5.23 released
by Michal Trojnara 02 Sep '15

02 Sep '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Dear Users, I have released version 5.23 of stunnel. The ChangeLog entry: Version 5.23, 2015.09.02, urgency: LOW * New features - Client-side support for the SOCKS protocol. See https://www.stunnel.org/socksvpn.html for details. - Reject SOCKS requests to connect loopback addresses. - New service-level option "OCSPnonce". The default value is "OCSPnonce = no". - Win32 directory structure rearranged. The installer script provides automatic migration for common setups. - Added Win32 installer option to install stunnel for the current user only. This feature does not deploy the NT service, but it also does not require aministrative privileges to install and configure stunnel. - stunnel.cnf was renamed to openssl.cnf in order to to prevent users from mixing it up with stunnel.conf. - Win32 desktop is automatically refreshed when the icon is created or removed. - The ca-certs.pem file is now updated on stunnel upgrade. - Inactive ports were removed from the PORTS file. - Added IPv6 support to the transparent proxy code. * Bugfixes - Compilation fix for OpenSSL version older than 1.0.0. - Compilation fix for mingw. Home page: https://www.stunnel.org/ Download: https://www.stunnel.org/downloads.html SHA-256 hashes: 0fc4a702afd2e857bae8ad1f39c51546eb282c3d0ff4f4d55d62aece7328ddea stunnel-5.23.tar.gz 40739ac4124bbe5192d7f68b020758295586ae65a16c1612829bffd1d3873363 stunnel-5.23-installer.exe 001e7d0a6d53b05b64fe62b0ca0623077d89f090b473e2cabed1eafa49196725 stunnel-5.23-android.zip Best regards, Mike -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAEBCAAGBQJV5q76AAoJEC78f/DUFuAUG54P/2ycqSOKbP3CN1MMKJj9NxBM 1uDuv/voF9OT/Y064X2bjy6dApo5HA9a4VHr1EaF+3x+Wofemhc5puLsgbfuFJKl nogN8cNdR2sk+GLooytfKp+jRZ8NAl2Rv5lLi/YE57RsMe91wI5kRxWKpeIDeG4L y7LhsCLmHNYkffsGSeOaXfc3rBBWOY39Lc78t3kezK14jIGZiCH/V32HRHMaWGiw 9oLBIRxGOVAscKcC2+MOGtXelKeJFrnAfW+v6qcZE1/SL4JnGHz8jBxOcvFbfrko 5yJfWYpgQg7NeLNovY+o1EXVxxw4Ty9ppLob9cRTGmx1Fgb6eE1uYVSAbx53lYKF 2z0J2MkKgMj07Mao5j45DiqRqa/kWDzKIN4sTCm5atDKB2ozrcfuDLLEU3W6XKrw 7XHPSs+FdkXQXYsfK7wbftLKfNEu8ojGKyMTMuKQcX/SkxyQuV6rYXA8gBm3HJCY dNp5lU5FdoiMv9S+uQpDL2j9GjOSNAV7qPzyRk8ivl1a1LonkbjQxHApvktSwkHt Gey6xweW5vt7FsKb15mFhx6ctBldM9lR7u4EacXWy+hq5UY6/5Rr/rbyCkCvG/NU Oq5LD2bqX22Xl4XeiVCW2gpLcqPWdbREY3YklGt22ln+PdyA+iGpxbeczUHjV5qX a6P76JpGk/pL5lhWfRV4 =i2gK -----END PGP SIGNATURE-----

1 0

(no subject)
by Eric Poythress 02 Sep '15

02 Sep '15

I am having some issues with getting Stunnel to work for my application. I'm using a logger that is able to email out alarms but it is not encrypted. I have Stunnel on the PC that connects to the logger. Both are able to see each other just fine but I have the following error in the log: Service [SMTP Outgoing] needs authentication to prevent MITM attacks I only want mail to go out and on the logger side of things my only inputs are SMTP Server, Use SMTP Authentication check box, Username, Password and Port. Any help or suggests would be greatly appreciated. Thank you, Eric Poythress Test Manager 1600 West Tacoma Street Broken Arrow, Oklahoma 74012 [Description: Air Hygiene Office - email signature] Direct - (918) 994-4171 Fax - (918) 307-9131 Mobile - (573) 587-0226 Toll Free - (888) 461-8778 virtual tour Air Hygiene: http://host.newspin360.net/airhygiene/ e-mail: epoythress(a)airhygiene.com<mailto:epoythress@airhygiene.com> website: www.airhygiene.com<http://www.airhygiene.com/>

2 3

Security suggestions on stunnel with SSL/TLS and SNI configuration
by Secure-Mail User 28 Aug '15

28 Aug '15

2 2

Re: [stunnel-users] stunk closing connections with RST
by Ade Ajibade 26 Aug '15

26 Aug '15

> > Hi All, > > I am having the same problem in my lab. > > I am trying to set up a secure connection between my Bluecoat proxy and stunnel running on windows. I need to encrypt the connection between the 2 devices. > > I am new to stunnel. > > My Bluecoat is configured to do health checks to the stunnel on the specified stunnel port. However every now and again, it reports the server is down. Looking at the capture, I can see the stunnel is sending RESETS back to the proxy. > > The TLS connection to the proxy appears to be working fine. But I need to be able to monitor the servers health. > > How do I do this? > > You have mentioned in your post the stunnel configured of client = yes. > > On what side does this need to be configured and what exactly does this mean? If it is configured as client = no what will be the implication. > > I have been looking at how you solved the F5 issue but how does that apply to what I am doing from my proxy. > > Please any help will be useful. I just subscribed to your site. > > Thank you > > > > Adekunle

1 0

New Windows installer
by Michal Trojnara 19 Aug '15

19 Aug '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dear Users, I have performed a major rewrite of the stunnel installer for Windows. It now allows installing stunnel for all users (as before), and also for the current user (in the user's home directory). It also no longer installs all the files into a single directory. The installer attempts to automatically migrate the previous layout into the new one. https://www.stunnel.org/downloads/beta/stunnel-5.23b4-installer.exe Please give it a try and let me know your comments and any bugs you can find. Mike -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJVziEpAAoJEC78f/DUFuAUFFIP/1CN9Bh5PFvpUlmnC67exLA4 CfPLXK4jj3PjFISF3scKAGJJjaCmyWQzF3mmTTn5HXprCfwOtZXf3jJQVI0YFZjE 8OgxpdcWhAyuOr2wQLSs5DX+WBILuxHBpIIJrjYIlrAsKMwmB7ucLHf17K+AkCjb j++ctJdFDQygpX0w9Qp0oyIbiEGlgaFHAn/aeYKdLSzTGaDkTptswMsM+SZzNcHF bK9VxQ15xFE6EwOp45roxc7eMuOd6VGQ9uyj1DcLaHygE0m5ALzXFN2Xsnii9UIq ZbzoOMBYlGYAya8Hbgrrb3D+53CwwQrhjVgE27FbMp3qc4ijboaBlfSzd+Bci3Rg B84EkJjBNC1cOcl5JvB/QAuJgXReBbFecS5+8PYNTzii6HU4bFXKb9JlbxCA7eGm HiCkn+DM4GJ3RXtWYjcyWfhlgCx69SUaEJvtphHl9QLbyGnDsJTE7rLlmpnp4B2N MNvZFhxuZrBsMmDRydQSYKaJ7S5e9H6soi630OfLxHDIbOGD+vDVvO9XrjtdYWBO 6Fm4ed0lDuTcqTzCxy93vbwPkWzvPyOykQPCED/UQ1RgtC8AIi6JVXhCwLs30pwj rSHgBKGxC5E19MXYwrNq3cr0nJdkw4oh8NShaXiplfhdCMdjRZHZz1HpbQzW0WP/ Z5BrUzv9yYpA/aAk4r7V =RSFJ -----END PGP SIGNATURE-----

2 3

net::ERR_TIMED_OUT error when connecting to a websocket via stunnel
by Mike Alhayek 17 Aug '15

17 Aug '15

Hi, I hope someone can help me here as I have been stuck for the past 3 days. I need to allow a client to connect to a websocket running on server 10.0.4.160 on port 8080. I need to setup a secure tunnel to allow the client to connect to the websocket via port 58443. I installed the latest version of stunnel on my windows Server 2008 R2. Then I changed my stunnel.conf file to look like this [websockets] client = yes accept = 58443 connect = 8080 verify = 2 checkIP = 10.0.4.195 CAfile = ca-certs.pem debug = 7 OCSPaia = yes ##The rest have been removed I am expecting to open a connection from the user's console on port 58443 and the stunnel will allow the client to connect to port 8080 This is the code that I use in the client's console "Of course, the console is on a PC that is on my internal network" var conn = new WebSocket('wss://10.0.4.160:58443'); conn.onopen = function(e) { console.log("Connection established!"); }; conn.onmessage = function(e) { console.log(e.data); }; While on the server, I open up a command line and I execute this telnet localhost 8080 to open up a command line where I can see the incoming request. As soon as I fire the code above from a client javascript's console I see this in the server's command line [cid:image001.png@01D0D8E6.8F1CC750] But few seconds later I get this error in the client's console WebSocket connection to 'wss://10.0.4.160:58443/' failed: Error in connection establishment: net::ERR_TIMED_OUT If I try to send a command using conn.sent('Hello') I get this error VM289:2 Uncaught DOMException: Failed to execute 'send' on 'WebSocket': Still in CONNECTING state. at Error (native) at <anonymous>:2:6 at Object.InjectedScript._evaluateOn (<anonymous>:905:140) at Object.InjectedScript._evaluateAndWrap (<anonymous>:838:34) at Object.InjectedScript.evaluate (<anonymous>:694:21) Additionally, I tried to turn on debugging in strunnel by adding debug = 7 to the config file This is what I got in the stunnel console. I don't understand what is wrong as the logs show that the client is connected but I can't send messages from the client to the server 2015.08.16 16:40:06 LOG7[36]: Service [websockets] started 2015.08.16 16:40:06 LOG5[36]: Service [websockets] accepted connection from 10.0.4.195:21963 2015.08.16 16:40:06 LOG6[36]: s_connect: connecting 127.0.0.1:8080 2015.08.16 16:40:06 LOG7[36]: s_connect: s_poll_wait 127.0.0.1:8080: waiting 10 seconds 2015.08.16 16:40:06 LOG5[36]: s_connect: connected 127.0.0.1:8080 2015.08.16 16:40:06 LOG5[36]: Service [websockets] connected remote server from 127.0.0.1:50891 2015.08.16 16:40:06 LOG7[36]: Remote socket (FD=668) initialized 2015.08.16 16:40:06 LOG6[36]: SNI: sending servername: localhost 2015.08.16 16:40:06 LOG7[36]: SSL state (connect): before/connect initialization 2015.08.16 16:40:06 LOG7[36]: SSL state (connect): SSLv2/v3 write client hello A What am I doing wrong here? How come the connection keeps failing?

1 0

Accept connection from sub-net?
by v v 12 Aug '15

12 Aug '15

Hi, Is it possible to configure stunnel server to accept connection from any IP in some subnet. For example, 111.22.33.xx. The proxy I connect through every time uses the different IP (from some IP pool) to connect to stunnel server. I don't want to add to the config file 255 possible host addresses. I don't see any suitable option in the stunnel documentation. Do I miss something? Any idea it could be done? Thanks, vov

1 0

Problem in connecting with root directory with stunnel connection in Mac OS
by Saurabh Beriwal 07 Aug '15

07 Aug '15

Hello All I am installing stunnel as client in Mac machine. I am using below configuration. accept=127.0.0.1:119 connect=target server information I am able to start stunnel service and its getting bound also. Now when I try to open the remote server directory with root folder , this does not work. However if I specify a shared folder name then this does work. I am facing this problem in Mac OS 10.10. smb://127.0.0.1:119/ does not work. smb://127.0.0.1:119/folderName works fine. However, both cases work fine in case of windows using stunnel. Please let me know if someone has faced similar issue in Mac OS with stunnel connection. Kindly help. Regards Saurabh Beriwal

1 0

STunnel nCipher openssl chil problems
by Marc Phillips 06 Aug '15

06 Aug '15

Has anyone managed to get STunnel to work with nCipher using the chil engine? STunnel appears to load the engine just fine, but failes when loading the key: [ ] Enabling support for engine "chil" [ ] Initializing engine #1 (chil) [ ] Engine #1 (chil) initialized [.] FIPS mode disabled [ ] Compression disabled [ ] PRNG seeded successfully [ ] Initializing service [https-server] [ ] Loading certificate from file: /usr/local/stunnel/etc/keystores/servercert.crt [ ] Loading key from engine: /usr/local/stunnel/etc/keystores/servercert.key [!] error queue: 26096080: error:26096080:engine routines:ENGINE_load_private_key:failed loading private key [!] ENGINE_load_private_key: 80069066: error:80069066:CHIL engine:HWCRHK_LOAD_PRIVKEY:chil error [!] Service [https-server]: Failed to initialize SSL context Using openssl directly, the key works fine (both with s_server and s_client). My stunnel config is pretty simple: pid = /usr/local/var/log/stunnel.pid debug = 7 output = /usr/local/var/log/stunnel.log options = -NO_SSLv3 engine = chil CApath = /usr/local/etc/cacerts/ [https-server] engineNum=1 accept = 4466 connect = 4433 cert=/usr/local/stunnel/etc/keystores/servercert.crt key=/usr/local/stunnel/etc/keystores/servercert.key R. Marc

1 0

Install Guide for STunnel Version 5.22
by Deepak Subbanarasimha 05 Aug '15

05 Aug '15

All, Is there an install guide for Stunnel version 5.22 for windows? If so, could you point me to it? -Deepak IMPORTANT NOTICE: This email is intended solely for the use of the individual to whom it is addressed and may contain information that is privileged, confidential or otherwise exempt from disclosure under applicable law. If the reader of this email is not the intended recipient or the employee or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution, or copying of this communication is strictly prohibited. If you have received this communication in error, please immediately return the original message to the sender at the listed email address. In accordance with Kewill policy, emails sent and received may be monitored. Although Kewill takes reasonable precautions to minimize the risk, Kewill accepts no responsibility for any loss or damage should this email contain any virus, or similar destructive or mischievous code.

2 2

"Chained" stunnel instances
by Michael Gebis 01 Aug '15

01 Aug '15

Hello. I have a question about a strange stunnel configuration; specifically, I'm like to use 'chained' stunnel instances, and I'm running into an issue. We have a conceptually simple setup: a client that connects to a server. We use stunnel both for encryption and for the failover mechanism. Here's a diagram of our simplest setup: /----S1 / C--st0-----S2 \ \----S3 We have a client that connects to stunnel. Our stunnel configuration lists three connections with "prio" failover mode. So usually, connections go from C thru st and onto Server 1. If S1 is down, st0 fails to connect to S1 and instead tries S2, and all is good. However, sometimes we may place an optional second instance of stunnel in front of the servers. /----st1--S1 / C--st0-----st2--S2 \ \----st3--S3 The failover mode of stunnel does not work so well in this configuration. If S1 is down, st0's failover algorithm does not kick in. Instead, st0 happily connects to st1, which is still alive and running. st1 then detects S1 is down and immediately closes the connection, but st0 does not care. Since the initial connection was successful, it does not initiate the failover algorithm. You may ask "why not change to round-robin mode?" The answer is that S1 is a dedicated machine, and S2/S3 are underpowered backups that have other primary responsibilities. We really want to direct all connections to S1 and only use S2/S3 in emergencies. You may also ask, "Why the second layer of stunnel?"--unfortunately, there are several hairy implementation-specific details that make this hard to change. My question is: is there any stunnel configuration option that can help us out? We would like the failover to work with and without the second layer of stunnel. From looking at the source code, I think I'm out of luck, but I figured it couldn't hurt to ask. Thanks! Michael

3 3