- stunnel-users - stunnel.org

Build failure for ARM32 with Yocto
by Laszlo Papp 22 Aug '13

22 Aug '13

Hi, if I build stunnel manually on my desktop machine (Archlinux), it works fine. However, when I am trying to use a Yocto/OE recipe for building it for my arm32 target, I am facing the issue below. Anyone any hint? Cheers, L. | /usr//include/linux/sysctl.h:836:20: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:849:12: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:892:31: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:897:32: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:912:26: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:917:32: warning: comma at end of enumerator list [-pedantic] | /usr//include/linux/sysctl.h:928:20: warning: comma at end of enumerator list [-pedantic] | In file included from /usr//include/linux/netfilter_ipv4.h:8:0, | from common.h:383, | from network.c:38: | /usr//include/linux/netfilter.h:58:18: warning: comma at end of enumerator list [-pedantic] | In file included from common.h:383:0, | from network.c:38: | /usr//include/linux/netfilter_ipv4.h:68:26: warning: comma at end of enumerator list [-pedantic] | make[2]: *** [stunnel-str.o] Error 1 | make[2]: *** Waiting for unfinished jobs.... | make[2]: *** [stunnel-network.o] Error 1 | make[2]: *** [stunnel-resolver.o] Error 1 | make[2]: *** [stunnel-protocol.o] Error 1 | make[2]: *** [stunnel-log.o] Error 1 | make[2]: *** [stunnel-file.o] Error 1 | make[2]: *** [stunnel-options.o] Error 1 | make[2]: *** [stunnel-client.o] Error 1 | make[2]: Leaving directory `/home/lpapp/Projects/Yocto/poky-dylan-9.0.1/build/tmp/work/armv5te-linux-gnueabi/stunnel/4.56-r0/stunnel-4.56/src' | make[1]: *** [all] Error 2 | make[1]: Leaving directory `/home/lpapp/Projects/Yocto/poky-dylan-9.0.1/build/tmp/work/armv5te-linux-gnueabi/stunnel/4.56-r0/stunnel-4.56/src' | make: *** [all-recursive] Error 1 | ERROR: oe_runmake failed | ERROR: Function failed: do_compile (see /home/lpapp/Projects/Yocto/poky-dylan-9.0.1/build/tmp/work/armv5te-linux-gnueabi/stunnel/4.56-r0/temp/log.do_compile.10590 for further information) ERROR: Task 6 (/home/lpapp/Projects/Yocto/poky-dylan-9.0.1/build/../meta-networking/recipes-support/stunnel/ stunnel_4.56.bb, do_compile) failed with exit code '1' NOTE: Tasks Summary: Attempted 561 tasks of which 559 didn't need to be rerun and 1 failed. No currently running tasks (561 of 568) Summary: 1 task failed: /home/lpapp/Projects/Yocto/poky-dylan-9.0.1/build/../meta-networking/recipes-support/stunnel/ stunnel_4.56.bb, do_compile Summary: There was 1 ERROR message shown, returning a non-zero exit code.

2 3

Problem Selecting Only Ecliptic Curve Ciphers
by Carter Browne 07 Aug '13

07 Aug '13

I'm using stunnel 4.56 on Windows 7. When I use the following cipher list: ciphers = ECDHE-ECDSA-AES256-SHA:ECDH-ECDSA-AES256-SHA:ECDHE-ECDSA-AES128-SHA:ECDH-ECDSA-AES128-SHA to establish a connection, I get a "no shared cipher" response. The following set of ciphers does work: ciphers = ECDHE-RSA-AES256-SHA:ECDHE-ECDSA-AES256-SHA:ECDH-RSA-AES256-SHA:ECDH-ECDSA-AES256-SHA:ECDHE-RSA-AES128-SHA:ECDHE-ECDSA-AES128-SHA:ECDH-RSA-AES128-SHA:ECDH-ECDSA-AES128-SHA Other relevant settings: options = NO_SSLv2 sslVersion = all fips = no verify = 2 If I take out the first ECDHE-RSA-ASE256-SHA cipher from the list, the ECDHE-RSA-AES128-SHA cipher is selected. What am I doing wrong? Thanks. -- Carter Browne cbrowne(a)cbcs-usa.com

2 1

Unable to make SSL connexion work
by OLIVER Patrice 07 Aug '13

07 Aug '13

Hi, Since our company had to migrate to Orange Office365 platform, we cannot send simple SMTP message using smtp.office365.com We have to use SMTP over TLS. While I have to use some applications which cannot use SMTP, I installed and setup STunnel. When I try to send a simple mail like this : telnet myserver 25 ehlo ch-lepuy.fr mail from: toto(a)ch-lepuy.fr I cannot continue with rcpt to syntax and I have the following message : rcpt to:530 5.7.1 Client was not authenticated In the STunnel log file, I can see that : 2013.08.06 19:20:49 LOG7[1352:2832]: SSL state (connect): before/connect initialization 2013.08.06 19:20:49 LOG7[1352:2832]: SSL state (connect): SSLv3 write client hello A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read server hello A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read server certificate A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read server key exchange A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read server certificate request A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read server done A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 write client certificate A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 write client key exchange A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 write certificate verify A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 write change cipher spec A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 write finished A 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 flush data 2013.08.06 19:20:50 LOG7[1352:2832]: SSL state (connect): SSLv3 read finished A 2013.08.06 19:20:50 LOG7[1352:2832]: 1 items in the session cache 2013.08.06 19:20:50 LOG7[1352:2832]: 1 client connects (SSL_connect()) 2013.08.06 19:20:50 LOG7[1352:2832]: 1 client connects that finished 2013.08.06 19:20:50 LOG7[1352:2832]: 0 client renegotiations requested 2013.08.06 19:20:50 LOG7[1352:2832]: 0 server connects (SSL_accept()) 2013.08.06 19:20:50 LOG7[1352:2832]: 0 server connects that finished 2013.08.06 19:20:50 LOG7[1352:2832]: 0 server renegotiations requested 2013.08.06 19:20:50 LOG7[1352:2832]: 0 session cache hits 2013.08.06 19:20:50 LOG7[1352:2832]: 0 external session cache hits 2013.08.06 19:20:50 LOG7[1352:2832]: 0 session cache misses 2013.08.06 19:20:50 LOG7[1352:2832]: 0 session cache timeouts 2013.08.06 19:20:50 LOG7[1352:2832]: Peer certificate was cached (6184 bytes) 2013.08.06 19:20:50 LOG6[1352:2832]: SSL connected: new session negotiated 2013.08.06 19:20:50 LOG6[1352:2832]: Negotiated TLSv1/SSLv3 ciphersuite: ECDHE-RSA-AES256-SHA (256-bit encryption) 2013.08.06 19:20:50 LOG6[1352:2832]: Compression: null, expansion: null 2013.08.06 19:21:13 LOG3[1352:2832]: SSL_read: Connection reset by peer (WSAECONNRESET) (10054) 2013.08.06 19:21:13 LOG5[1352:2832]: Connection reset: 65 byte(s) sent to SSL, 231 byte(s) sent to socket 2013.08.06 19:21:13 LOG7[1352:2832]: Remote socket (FD=316) closed 2013.08.06 19:21:13 LOG7[1352:2832]: Local socket (FD=244) closed 2013.08.06 19:21:13 LOG7[1352:2832]: Service [SMTP Outgoing] finished (0 left) Here is my STunnel setup : client = yes output = stunnel-log.txt debug = 7 taskbar = yes ; Disable FIPS mode to allow non-approved protocols and algorithms ;fips = no ; ************************************************************************** ; * Service defaults may also be specified in individual service sections * ; ************************************************************************** ; Certificate/key is needed in server mode and optional in client mode cert = stunnel.pem ; Disable support for insecure SSLv2 protocol options = NO_SSLv2 [SMTP Outgoing] protocol = smtp accept = 25 connect = smtp.office365.com:587 Does anyone have an idea ? Regards? Patrice Oliver | Centre Hospitalier Emile Roux Responsable Informatique T. 04 71 04 38 30 12 Bd Andr? Chantemesse - 43012 LE PUY EN VELAY Cedex http://www.ch-lepuy.fr

3 3

Disable SSL session ticket
by dansmith 06 Aug '13

06 Aug '13

Hi, when connecting a stunnel client to a stunnel server, is it possible to disable the use of SSL session tickets? I need to be able to decrypt the stunnel traffic in wireshark. Unfortunately, wireshark still doesn't support SSL session tickets.

2 1

transparent = source, not working
by Ruben Cardenal 04 Aug '13

04 Aug '13

Hi, I'm trying to setup yet another service of this kind. I've seen this has been largely discussed several times on this list (but without valid solutions), and I'm writting this email as some kind of last resort after hours of testing and debugging. Pretty simple configuration: # cat /etc/stunnel/stunnel.conf cert = /etc/ssl/certs/stunnel4/my-cert.crt key = /etc/ssl/certs/stunnel4/my-cert.key sslVersion = SSLv3 foreground = yes pid = /tmp/stunnel4.pid socket = l:TCP_NODELAY=1 socket = r:TCP_NODELAY=1 debug = 7 output = /var/log/stunnel.log [service] accept = 195.78.X.X:6697 connect = 195.78.X.X:1357 transparent = source Accept and connect IP's are the same. # /usr/local/bin/stunnel -version stunnel 4.56 on i686-pc-linux-gnu platform Compiled/running with OpenSSL 0.9.8o 01 Jun 2010 Threading:PTHREAD Sockets:POLL,IPv6 SSL:ENGINE,OCSP Global options: debug = daemon.notice pid = /usr/local/var/run/stunnel/stunnel.pid RNDbytes = 64 RNDfile = /dev/urandom RNDoverwrite = yes Service-level options: ciphers = ALL:!SSLv2:!aNULL:!EXP:!LOW:-MEDIUM:RC4:+HIGH curve = prime256v1 sessionCacheSize = 1000 sessionCacheTimeout = 300 seconds sslVersion = TLSv1 for client, all for server stack = 65536 bytes TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none # /usr/local/bin/stunnel -sockets stunnel 4.56 on i686-pc-linux-gnu platform Compiled/running with OpenSSL 0.9.8o 01 Jun 2010 Threading:PTHREAD Sockets:POLL,IPv6 SSL:ENGINE,OCSP Socket option defaults: Option Name | Accept | Local | Remote |OS default ----------------+----------+----------+----------+---------- SO_DEBUG | -- | -- | -- | no SO_DONTROUTE | -- | -- | -- | no SO_KEEPALIVE | -- | -- | -- | no SO_LINGER | -- | -- | -- | 0:0 SO_OOBINLINE | -- | -- | -- | no SO_RCVBUF | -- | -- | -- | 87380 SO_SNDBUF | -- | -- | -- | 65536 SO_RCVLOWAT | -- | -- | -- | 1 SO_SNDLOWAT | -- | -- | -- | 1 SO_RCVTIMEO | -- | -- | -- | 0:0 SO_SNDTIMEO | -- | -- | -- | 0:0 SO_REUSEADDR | yes| -- | -- | no SO_BINDTODEVICE | -- | -- | -- |write-only TCP_KEEPCNT | -- | -- | -- | 9 TCP_KEEPIDLE | -- | -- | -- | 7200 TCP_KEEPINTVL | -- | -- | -- | 75 IP_TOS | -- | -- | -- | 0 IP_TTL | -- | -- | -- | 64 TCP_NODELAY | -- | yes| yes| no IP_FREEBIND | -- | -- | -- | no And the timeout, the same I've seen suffering to other people: 2013.08.03 19:52:12 LOG7[18496:3074533056]: Service [service] accepted (FD=3) from MY_HOME_ADDRESS:34836 2013.08.03 19:52:12 LOG7[18496:3074530160]: Service [service] started 2013.08.03 19:52:12 LOG5[18496:3074530160]: Service [service] accepted connection from MY_HOME_ADDRESS:34836 (blah blah ssl stuff) 2013.08.03 19:52:12 LOG6[18496:3074530160]: Negotiated TLSv1/SSLv3 ciphersuite: DHE-RSA-AES256-SHA (256-bit encryption) 2013.08.03 19:52:12 LOG6[18496:3074530160]: Compression: null, expansion: null 2013.08.03 19:52:12 LOG6[18496:3074530160]: IP_TRANSPARENT socket option set 2013.08.03 19:52:12 LOG6[18496:3074530160]: local_bind succeeded on the original port 2013.08.03 19:52:12 LOG6[18496:3074530160]: connect_blocking: connecting 195.78.X.X:1357 2013.08.03 19:52:12 LOG7[18496:3074530160]: connect_blocking: s_poll_wait 195.78.X.X:1357: waiting 60 seconds 2013.08.03 19:52:21 LOG3[18496:3074530160]: connect_blocking: connect 195.78.X.X:1357: Connection timed out (110) 2013.08.03 19:52:21 LOG5[18496:3074530160]: Connection reset: 0 byte(s) sent to SSL, 0 byte(s) sent to socket iptables stuff and that, is in place: # cat /proc/sys/net/ipv4/conf/all/rp_filter 0 # cat /proc/sys/net/ipv4/ip_forward 1 And did the iptables part: iptables -t mangle -N DIVERT iptables -t mangle -A PREROUTING -p tcp -m socket -j DIVERT iptables -t mangle -A DIVERT -j MARK --set-mark 1 iptables -t mangle -A DIVERT -j ACCEPT ip rule add fwmark 1 lookup 100 ip route add local 0.0.0.0/0 dev lo table 100 Debugging the whole thing, it can be seen that stunnel tries to connect: [pid 16823] connect(9, {sa_family=AF_INET, sin_port=htons(1357), sin_addr=inet_addr("195.78.X.X")}, 16) = -1 EINPROGRESS (Operation now in progress) BUT the service running in 1357 does this: # tcpdump -i eth1 -n port 1357 tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth1, link-type EN10MB (Ethernet), capture size 65535 bytes 19:52:52.586773 IP 195.78.X.X.1357 > MY_HOME_ADDRESS.34853: Flags [S.], seq 2655966098, ack 546202865, win 5840, options [mss 1460,nop,nop,sackOK], length 0 And, according to that, it looks obvious to me that this setup will never work, since that ACK packet goes to my home box, and not to the local connection. So either I'm doing something wrong (I hope I am!!) or this thing definitely doesn't work... Any help/ideas/whatever, please? Thanks, Rubén.

2 5

Library for iOS
by ios dev 29 Jul '13

29 Jul '13

Hi. I have to build static library for using in iOS application. Did anybody already built it? How can I do it? Thanks.

1 0

Reverse DNS lookup in stunnel log possible?
by mkanet＠yahoo.com 28 Jul '13

28 Jul '13

I currently have stunnel strip SSL from incoming https connections; which then passes the connections to a proxy before ultimately reaching my web server. So, the only easy way to see where incoming connections are coming from are in the stunnel log. Below, is a small example of what my stunnel log looks like (no, those arent the real IPs ). The information below would be much more useful to me if it included the DNS names in addition to their numeric IP. I currently have the latest Windows version of stunnel installed. It would be great to know how to get it to resolve DNS names as well in the log file; preferably without impeding general stunnel performance. I tried several debug levels; but none them did reverse DNS lookup. Hopefully someone know how to do this on a Windows stunnel setup. 2013.07.23 10:16:00 LOG5[10152:15136]: Service [stunnel-sslh] connected remote server from 24.12.152.129:58773 2013.07.23 10:16:00 LOG3[10152:15136]: SSL_read: Connection reset by peer (WSAECONNRESET) (10054) 2013.07.23 10:16:00 LOG5[10152:15136]: Connection reset: 272 byte(s) sent to SSL, 96 byte(s) sent to socket 2013.07.23 10:17:53 LOG5[10152:4000]: Service [stunnel-sslh] accepted connection from 71.194.51.232:5535 2013.07.23 10:17:53 LOG5[10152:4000]: connect_blocking: connected 24.12.152.129:7777 2013.07.23 10:17:53 LOG5[10152:4000]: Service [stunnel-sslh] connected remote server from 24.12.152.129:58799 2013.07.23 10:17:53 LOG5[10152:13212]: Service [stunnel-sslh] accepted connection from 71.194.51.232:5508 2013.07.23 10:17:53 LOG5[10152:3348]: Service [stunnel-sslh] accepted connection from 71.194.51.232:5509 2013.07.23 10:17:53 LOG5[10152:2884]: Service [stunnel-sslh] accepted connection from 71.194.51.232:5519

4 9

Re: [stunnel-users] Server down
by Ludolf Holzheid 24 Jul '13

24 Jul '13

On Wed, 2013-07-24 13:22:26 -0400, Walter Hankins wrote: > Hi Ludoff, > > I am new to stunnel. I trying to configure. > > Stunnel.conf: > socket = l: TCP_NODELAY = 1 > socket = r: TCP_NODELAY = 1 > > debug = 7 > output = stunnel.log Hi Walter, Always reply to the list, so others can comment. Your log says `Cannot open log file: stunnel.log´. I don't how you invoke stunnel, so I don't know if the directory it's working in is writable, but you may try to specify an output file in a writable directory with an absolute path, e.g. output=C:/stunnel.log if C:\ is writable on your box. HTH, Ludolf -- Bihl+Wiedemann GmbH Floßwörthstraße 41 68199 Mannheim, Germany Tel: +49 621 33996-0 Fax: +49 621 3392239 mailto:[email protected] http://www.bihl-wiedemann.de Sitz der Gesellschaft: Mannheim Geschäftsführer: Jochen Bihl, Bernhard Wiedemann Amtsgericht Mannheim, HRB 5796

1 0

Server down
by Walter Hankins 24 Jul '13

24 Jul '13

Server is down Reading configuration from file stunnel.conf FIPS mode is enabled Compression not enabled Snagged 64 random bytes from C:/.rnd Wrote 0 new random bytes to C:/.rnd PRNG seeded successfully Initializing service [Ssmtp] Certificate: stunnel.pem Certificate loaded Key file: stunnel.pem Private key loaded SSL options set: 0x01000004 Configuration successful Service [Ssmtp] (FD=672) bound to 127.0.0.1:25 Cannot open log file: stunnel.log Closing service [Ssmtp] Service [Ssmtp] closed (FD=672) Sessions cached before flush: 0 Sessions cached after flush: 0 Service [Ssmtp] closed Server is down

2 1

Re: [stunnel-users] stunnel and keepalive
by David H. Durgee 23 Jul '13

23 Jul '13

> `stunnel -sockets´ should show the available socket options and their > default values. > > For the box I'm writing this mail from, the default for TCP_KEEPIDLE > is 7200 and for TCP_KEEPINTVL 75 (both in seconds). You'll have to be > quite patient if you want to watch keepalive packets in wireshark > without tweaking the defaults ... > > HTH, > > Ludolf You hit the nail on the head. I set the parameters to something more reasonable and I now see the keepalive packets. Thank you for your speedy response. Dave

1 0

keepalive and stunnel
by David H. Durgee 23 Jul '13

23 Jul '13

I am attempting to use stunnel to provide secure telnet access to a server on my system. As it is possible that a user may be idle for long period I have attempted to enable keepalive with stunnel. My stunnel conf files contain: socket = l:SO_KEEPALIVE=1 socket = r:SO_KEEPALIVE=1 on both the server and client side. But wireshark shows no keepalive packets being exchanged. I am using socat to provide a service on the LAN and connections using socat do show keepalive packets. But to get this working with socat I had to use the following: socat -d -d -L/var/lock/Port$PORT - TCP:127.0.0.1:$PORT,keepalive,keepcnt=10,keepidle=3,keepintvl=1 I had originally tried it with only the keepalive option, but it took adding the other three options for it to work. Perhaps this is particular to my linux mint 13 maya x64 system and keepalive works by itself on windows? Are the other three options supported in stunnel? Dave

2 1

What do I need to provide an SSL service and verify individual clients?
by John Long 22 Jul '13

22 Jul '13

Hi, I want to set up a service using stunnel as a server. I want only specific clients to be able to connect. Each client will have an SSL-enabled client app. I understand the part about needing to create a self-signed cert for the stunnel server. I don't understand what Stunnel will require from each client. Do I have to create CSRs for each client and sign them with my self-signed cert, or will the keys from additional standalone self-signed certs for each client be good enough on the client side as long as I have a cert that for each key in the certificate path of the server? Is it correct that verify=3 will make sure only clients that have keys matching the certs in the server cert path can connect? If a client with a key that I have no cert for tries to connect, what should happen? Will it time out or will there be an error that the key isn't valid? Thanks, /jl

1 0

Stunnel log file limit
by mehmet ozisik 18 Jul '13

18 Jul '13

Hi, I would like to ask some questions regarding stunnel logging which I could not find answer with googling. 1) what is the limit of stunnel log file (can we define this)? 2) If there is a limit, when the limit is reached does it remove the filled file and starts with the same name an empty log file? Regards Mehmet

3 5

Feature requests
by Death Master 17 Jul '13

17 Jul '13

Can you implement some features? 1) Load balancing leastconn (like leastconn in haproxy) 2) UDT transport between stunnel instances(with several congestion controls which are already implemented for UDT) 3) SCTP transport between stunnel instances (if can be done in userspace) P.S. I'm using 400mbit WAN with 55-90ms latency+0-10%packet loss and on one end I can't control TCP settings at all (remote end is on OpenVZ VPS).

2 1

Line 51: "[pop3s]": Failed to initialize SSL
by Laura Fried 15 Jul '13

15 Jul '13

Beginning with Stunnel version 4.55 and continuing with 4.56 I can configure, make and make install without issue. But as soon as I try to start stunnel with the following command: /usr/local/bin/stunnel /usr/local/etc/stunnel/stunnel.conf I get the folllowing error message: Clients allowed=500 stunnel 4.56 on x86_64-unknown-linux-gnu platform Compiled/running with OpenSSL 1.0.1 14 Mar 2012 Threading:PTHREAD Sockets:POLL,IPv6 SSL:ENGINE,OCSP,FIPS Reading configuration from file /usr/local/etc/stunnel/stunnel.conf FIPS_mode_set: F06D065: error:0F06D065:common libcrypto routines:FIPS_mode_set:fips mode not supported Line 51: "[pop3s]": Failed to initialize SSL str_stats: 9 block(s), 163 data byte(s), 522 control byte(s) Any insight into this issue would be greatly appreciated. -- Laura J. Fried

3 5

is verify level 4 working?
by dansmith 11 Jul '13

11 Jul '13

I would expect that level 4 only compares locally installed certificates, however I get the same behaviour as with level 3, stunnel expects a CA cert. Here'e the relevant log when on level 4 Jul 6 23:46:31 mmm stunnel: LOG7[7870:140491349628672]: Starting certificate verification: depth=0, /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: CERT: Verification error: unable to get local issuer certificate Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: Certificate check failed: depth=0, /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd Jul 6 23:46:31 mmm stunnel: LOG7[7872:140080853112576]: SSL alert (read): fatal: unknown CA What am I missing in understanding verify's level 4 ?

3 6

Creating a Centralized Secure Log Server with syslog-ng and Stunnel
by Szajt, Pablo 10 Jul '13

10 Jul '13

Hi guys, I have the task to install a secure syslog server using syslog-ng and stunnel. The Syslog server is a Solaris 10 and Syslog clients are HP-UX 11.31. I've spent already so many hours and I cannot make it run properly. I know syslog-ng works well without stunnel. I'm pretty sure someone is already using it. I'd like to have step by step installation process. I have no issue if I have to start from scratch. Thanks for your help, Pablo Pablo Szajt

2 1

Stunnel 4.56 and FIPS
by Ondrej Hrebicek 09 Jul '13

09 Jul '13

Hello stunnel users, I'm trying to compile stunnel 4.56 with FIPS support on Ubuntu 12.04. Always end up with the infamous "fingerprint does not match" error. I can't figure out what I'm doing wrong. 1. Download openssl-fips-2.0.2.tar.gz, unpack, ./config, make, and sudo make install (as specified in http://www.openssl.org/docs/fips/UserGuide-2.0.pdf) 2. Download openssl-1.0.1e.tar.gz, unpack, ./config fips --with-fipslibdir=/usr/local/ssl/fips-2.0/lib/ --with-fipsdir=/usr/local/ssl/fips-2.0/, make depend, make, and sudo make install 3. At this point, openssl is working in FIPS mode: > OPENSSL_FIPS=1 /usr/local/ssl/bin/openssl version OpenSSL 1.0.1e-fips 11 Feb 2013 > OPENSSL_FIPS=1 /usr/local/ssl/bin/openssl sha1 c_rehash SHA1(c_rehash)= 5af9e1479950bbbd9d3304c181b3f802c54f64fd > OPENSSL_FIPS=1 /usr/local/ssl/bin/openssl md5 c_rehash Error setting digest md5 139806582736544:error:060A80A3:digital envelope routines:FIPS_DIGESTINIT:disabled for fips:fips_md.c:180: 4. Download stunnel-4.56.tar.gz, unpack, ./configure --enable-fips --with-ssl=/usr/local/ssl, make, and sudo make install 5. While configuring and building stunnel completes as expected, the following does appear in ./configure's output: checking whether to enable FIPS mode support... yes configure: **************************************** SSL checking for SSL directory... /usr/local/ssl checking /usr/local/ssl/include/openssl/engine.h usability... yes checking /usr/local/ssl/include/openssl/engine.h presence... yes checking for /usr/local/ssl/include/openssl/engine.h... yes checking /usr/local/ssl/include/openssl/ocsp.h usability... yes checking /usr/local/ssl/include/openssl/ocsp.h presence... yes checking for /usr/local/ssl/include/openssl/ocsp.h... yes checking /usr/local/ssl/include/openssl/fips.h usability... no checking /usr/local/ssl/include/openssl/fips.h presence... no checking for /usr/local/ssl/include/openssl/fips.h... no configure: WARNING: OpenSSL fips header not found This is not entirely unexpected as fips.h only exists in /usr/local/ssl/fips-2.0/include/openssl. 6. Running stunnel however fails: > /usr/local/bin/stunnel Clients allowed=500 stunnel 4.56 on x86_64-unknown-linux-gnu platform Compiled/running with OpenSSL 1.0.1e-fips 11 Feb 2013 Threading:PTHREAD Sockets:POLL,IPv6 SSL:ENGINE,OCSP,FIPS Reading configuration from file /usr/local/etc/stunnel/stunnel.conf FIPS_mode_set: 2D06B06F: error:2D06B06F:FIPS routines:FIPS_check_incore_fingerprint:fingerprint does not match Line 61: "[pop3s]": Failed to initialize SSL str_stats: 5 block(s), 120 data byte(s), 290 control byte(s) I can't figure out what's causing this, hoping someone on the list may have a couple suggestions. Thanks in advance!

2 2

Re: [stunnel-users] is verify level 4 working?
by Thomas Eifert 08 Jul '13

08 Jul '13

dansmith, I'd be the last to argue with you. Purely as an experiment, I tried changing the verify level from 4 to 3, but it didn't fly. I most likely missed the entirety of the signing chain, as I only added the certificate of the *issuing* CA. Regards, Thomas On 7/8/2013 7:01 PM, dansmith wrote: > Thomas, the recent exception you are describing - needing a CA > sertificate and server certificate is what verify=3 does. > So, I guess there is some regression in the code. > When I look at stunnel's verify.c code, there is only one reference to > level 4 in line 225. It seems like verify=4 functionality is missing > from the code. > > On 07/08/2013 11:44 PM, Thomas Eifert wrote: >> dansmith, >> >> It's my understanding that verify = 4 should, theoretically, look only >> for the server certificate, and this is the way I've been using it >> with great success over the past year or so. Recently, however, I ran >> into an exception to that behavior. >> >> In my case, I only had to download and install one certificate; that >> of the signing CA. I simply pasted it directly below the server >> certificate in the associated .pem file. The CA certificate wasn't >> originally in .pem format, so I converted it beforehand. OpenSSL has >> conversion capability, and there are also online certificate tools >> available. Your mileage may vary. >> >> Good luck. >> >> Thomas >> >> >> On 7/8/2013 3:01 PM, dansmith wrote: >> >>> Could you kindly break it down for me. Are you saying that I need to >>> have two CAs A & B. A signs the certificate of B and B signs the >>> certificate of my server? >>> Do I understand correctly that verify=4 is supposed to simply ignore >> any >>> CAs and only look at the actual certificate, comparing it to the >>> certificate in CAfile ? >>> >>> >>> On 07/08/2013 06:32 PM, Thomas Eifert wrote: >>>> You're not missing anything. I've experienced a similar issue. While >>>> verify = 4 generally works well in most cases and will ignore the CA >>>> chain, I've encountered a few isolated incidences in which I've had to >>>> append or "chain" the server certificate with the certificate of the >>>> CA. Give it a shot and see if it resolves your issue. >>>> >>>> Thomas >>>> >>>> On 7/8/2013 3:02 AM, dansmith wrote: >>>>> I would expect that level 4 only compares locally installed >>>>> certificates, however I get the same behaviour as with level 3, >> stunnel >>>>> expects a CA cert. >>>>> Here'e the relevant log when on level 4 >>>>> >>>>> Jul 6 23:46:31 mmm stunnel: LOG7[7870:140491349628672]: Starting >>>>> certificate verification: depth=0, >>>>> /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd >>>>> Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: CERT: >>>>> Verification error: unable to get local issuer certificate >>>>> Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: Certificate >>>>> check failed: depth=0, >> /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd >>>>> Jul 6 23:46:31 mmm stunnel: LOG7[7872:140080853112576]: SSL alert >>>>> (read): fatal: unknown CA >>>>> >>>>> What am I missing in understanding verify's level 4 ? >>>>> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> stunnel-users mailing list >>>>> stunnel-users(a)stunnel.org >>>>> https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users >>>>> >>>> >>> >>> >> > > -- Attention: This message and all attachments are private and may contain information that is confidential and privileged. If you received this message in error, please notify the sender by reply email and delete the message immediately.

1 0

is verify level 4 working?
by Thomas Eifert 08 Jul '13

08 Jul '13

dansmith, It's my understanding that verify = 4 should, theoretically, look only for the server certificate, and this is the way I've been using it with great success over the past year or so. Recently, however, I ran into an exception to that behavior. In my case, I only had to download and install one certificate; that of the signing CA. I simply pasted it directly below the server certificate in the associated .pem file. The CA certificate wasn't originally in .pem format, so I converted it beforehand. OpenSSL has conversion capability, and there are also online certificate tools available. Your mileage may vary. Good luck. Thomas On 7/8/2013 3:01 PM, dansmith wrote: > Could you kindly break it down for me. Are you saying that I need to > have two CAs A & B. A signs the certificate of B and B signs the > certificate of my server? > Do I understand correctly that verify=4 is supposed to simply ignore any > CAs and only look at the actual certificate, comparing it to the > certificate in CAfile ? > > > On 07/08/2013 06:32 PM, Thomas Eifert wrote: >> You're not missing anything. I've experienced a similar issue. While >> verify = 4 generally works well in most cases and will ignore the CA >> chain, I've encountered a few isolated incidences in which I've had to >> append or "chain" the server certificate with the certificate of the >> CA. Give it a shot and see if it resolves your issue. >> >> Thomas >> >> On 7/8/2013 3:02 AM, dansmith wrote: >>> I would expect that level 4 only compares locally installed >>> certificates, however I get the same behaviour as with level 3, stunnel >>> expects a CA cert. >>> Here'e the relevant log when on level 4 >>> >>> Jul 6 23:46:31 mmm stunnel: LOG7[7870:140491349628672]: Starting >>> certificate verification: depth=0, >>> /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd >>> Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: CERT: >>> Verification error: unable to get local issuer certificate >>> Jul 6 23:46:31 mmm stunnel: LOG4[7870:140491349628672]: Certificate >>> check failed: depth=0, /C=qq/ST=qq/O=qqq/OU=rer/CN=redf/emailAddress=rfd >>> Jul 6 23:46:31 mmm stunnel: LOG7[7872:140080853112576]: SSL alert >>> (read): fatal: unknown CA >>> >>> What am I missing in understanding verify's level 4 ? >>> >>> >>> >>> >>> _______________________________________________ >>> stunnel-users mailing list >>> stunnel-users(a)stunnel.org >>> https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users >>> >> > > -- Attention: This message and all attachments are private and may contain information that is confidential and privileged. If you received this message in error, please notify the sender by reply email and delete the message immediately. -- Attention: This message and all attachments are private and may contain information that is confidential and privileged. If you received this message in error, please notify the sender by reply email and delete the message immediately.

1 0

Re: [stunnel-users] accept : too many open files error(24)
by Senthil Naidu 01 Jul '13

01 Jul '13

Hi, We are using 64 bit version of centos, and the ram of this system is 1gb so should we increase this limit to 100000 in the limits.conf Regards senthil On Sun, Jun 30, 2013 at 12:31 AM, Michael Curran <mike_curran(a)hotmail.com>wrote: > Are you running 32-kernel for centos? > > 32-bit limits in the 4500 range for open connections due to the memory > constraints > > We usually set our open file limits to 100I specifically for Stunnel user > > *Sent from my Verizon Wireless 4G LTE DROID* > > > Senthil Naidu <senthil.naidu(a)gmail.com> wrote: > > Hi > > > > We are using stunnel 4.32 on centos and our ulimits in limits.conf are as > follows > > > > * soft nofile 16384 > > * hard nofile 65536 > > > > We keep getting the error “too many open files(24)” and stunnel does not > accept new connection till we restart the same > > > > Is this ulimit proper or we need to increase the same , we are look for > 3000 to 5000 connections > > > > Regards > > Senthil >

2 2

accept : too many open files error(24)
by Senthil Naidu 29 Jun '13

29 Jun '13

Hi We are using stunnel 4.32 on centos and our ulimits in limits.conf are as follows * soft nofile 16384 * hard nofile 65536 We keep getting the error “too many open files(24)” and stunnel does not accept new connection till we restart the same Is this ulimit proper or we need to increase the same , we are look for 3000 to 5000 connections Regards Senthil

1 0

Re: [stunnel-users] Sending Email Without Authentication with gmail
by Stephen Gates 28 Jun '13

28 Jun '13

I downloaded the old argosoft freeware and it works! Praise God! Thanks to all for your time, help and suggestions!! Much appreciated. Stephen Gates Harold K. Scholz Company 7800 Serum Avenue Ralston, NE 68127 402-339-7600

2 1

Re: [stunnel-users] Sending Email from PLC Without Authentication with gmail - static ip/argosoft/php script
by Stephen Gates 28 Jun '13

28 Jun '13

> The only limitation I can recall on this freeware version is in the file size (5MB) and in the retries between relay errors. > > But you have to be aware when using a mail server that, thanks to Spamhaus, you need a static IP or almost all SMTPs out there will block your server to accept mail from (including Gmail)." > The router that my network is connected to has a static IP as far as I can tell and so does the PC and PLC on the LAN. How can I determine for sure if I have a static IP? I will try the Argosoft link. > Question? What does this device have available for resources. I use a php script that checks server status emails via gmail when there are changes in status, and it handles the connection directly to gmail. > I'm not sure. I don't know what php script is. I am about to try emailrelay from sourceforge.net. Anybody have experience with that? http://emailrelay.sourceforge.net/ Thanks, Stephen Gates Harold K. Scholz Company 7800 Serum Avenue Ralston, NE 68127 402-339-7600

1 0

Re: [stunnel-users] Sending Email from PLC Without Authentication with gmail
by Stephen Gates 27 Jun '13

27 Jun '13

"The issue with your situation is not stunnel but the fact that your SMTP client does not have the capability to provide authentication. As you mentioned, Gmail will require authentication information from the client. If that information is not provided by the SMTP client application, then the SMTP transaction will fail. ? I would suggest you consider setting up some kind of SMTP Relay. Which will serve as a proxy between your PLC and Google SMTP servers. ? You can also take a look at this document and if possible implement the last approach? http://support.google.com/a/bin/answer.py?hl=en <http://support.google.com/a/bin/answer.py?hl=en&answer=176600> &answer=176600" Thanks for your help Leandro. Do you have a suggestion for a SMTP Relay?? I have tried Free SMTP Server which was the only thing that has worked so far and worked almost perfectly except that it can only send out 10 emails a day. And I need more capacity for sending than that. I also have used trial versions of ArgoSoft, 1st Mail Server and QK SMTP Server but couldn't get those to work. I have tried the Google suggestion for using asmpx.l.google.com but I need to send to vtext.com addresses which are outside of the google domain. I tried adding the vtext.com addresses to googlegroups.com and sending emails using the asmpx.l.google.com server but vtext did not receive the emails for googlegroups.com. So I'm desperate to find some sort of solution. Thanks, Stephen Gates Harold K. Scholz Company 7800 Serum Avenue Ralston, NE 68127 402-339-7600

3 2

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

stunnel-users