- stunnel-users - stunnel.org

Timeout issues
by gorf＠hyperfunk.club 05 Aug '20

05 Aug '20

Hi, I'm using stunnel v5.56 on FreeBSD to proxy connections to a web server (running Apache), for a small number of users. Users connecting to the service keep running into issues with timeouts. I see timeout errors in the stunnel log messages: services stunnel[7510]: LOG3[188]: transfer: s_poll_wait: TIMEOUTclose exceeded: closing I tried changing stunnel to use the proxy protocol and enabled it on Apache, too. I don't know if that will help but at least then the web server gets the correct, remote IP address. Can anyone offer ideas/suggestions to fix this problem? Is there anything I can test or check to help diagnose the issue? The configuration is very basic: [apache] accept = 20888 protocol = proxy connect = 192.168.0.10:8000 PSKsecrets = /usr/local/etc/psksecrets.txt

1 0

Getting incoming Exchange online (Office 365) emails to go to port 110 and not 995
by Brent Kimberley 23 Jul '20

23 Jul '20

Office365 IMAP4S + POP3S is dropping legacy authentication in favour of OAUTH2.0 authentication -- effective October 2020. (Some may have longer.) How do you you plan to migrate to authentication? Do you plan to upgrade your client(s) or inject an OAUTH2.0 transparent proxy ? ---------------------------------------------------------------------- Message: 1 Date: Thu, 23 Jul 2020 01:22:06 +0000 From: Mel Gilbert <m.gilbert(a)baymedicalgroup.com.au> To: "stunnel-users(a)stunnel.org" <stunnel-users(a)stunnel.org> Subject: Re: [stunnel-users] Getting incoming Exchange online (Office 365) emails to go to port 110 and not 995 Message-ID: <SY4P282MB074766F32238B92A3D951499C1760(a)SY4P282MB0747.AUSP282.PROD.OUTLOOK.COM> Content-Type: text/plain; charset="us-ascii" Hi Trying to get an application with inbuilt email to receive emails using POP and port 110. However we use Office 365 and I have tried below config but it is not working. I want people to send email to an Exchange Online Account port 995 and it gets forwarded on to the email app on port 110 client = yes output = stunnel-log.txt debug=7 taskbar=yes [POP3 Incoming] accept = 110 connect = Outlook.office365.com:995 cert = stunnel.pem [SMTP Outgoing] protocol = smtp accept = 25 connect = smtp.office365.com:587 cert = stunnel.pem The outgoing email works fine using SMTP relay that I've set up on Office 365 but cant get the incoming emails to work. Any help would be appreciated Thanks Mel

1 0

Re: [stunnel-users] Getting incoming Exchange online (Office 365) emails to go to port 110 and not 995
by Mel Gilbert 23 Jul '20

23 Jul '20

Hi Trying to get an application with inbuilt email to receive emails using POP and port 110. However we use Office 365 and I have tried below config but it is not working. I want people to send email to an Exchange Online Account port 995 and it gets forwarded on to the email app on port 110 client = yes output = stunnel-log.txt debug=7 taskbar=yes [POP3 Incoming] accept = 110 connect = Outlook.office365.com:995 cert = stunnel.pem [SMTP Outgoing] protocol = smtp accept = 25 connect = smtp.office365.com:587 cert = stunnel.pem The outgoing email works fine using SMTP relay that I've set up on Office 365 but cant get the incoming emails to work. Any help would be appreciated Thanks Mel

1 0

Transparent Proxy help.
by Steven Relf 21 Jul '20

21 Jul '20

Hello List, I am working on a small project where I need to encrypt nfs traffic, and Stunnel looks to be ideal, the only issue I'm having is getting the transparent part to work I have a client machine, running stunnel config below #GLOBAL####################################################### sslVersion = TLSv1.2 TIMEOUTidle = 600 renegotiation = no FIPS = no options = NO_SSLv2 options = NO_SSLv3 options = SINGLE_DH_USE options = SINGLE_ECDH_USE options = CIPHER_SERVER_PREFERENCE syslog = yes debug = 7 ;chroot = /var/empty/stunnel libwrap = yes service = 3d-nfsd curve = secp521r1 #CREDENTIALS################################################## verify = 4 CAfile = /etc/stunnel/nfs-tls.pem cert = /etc/stunnel/nfs-tls.pem #ROLE######################################################### client = yes connect = fqdn:2363 and the client which is running ontop of the nfs-ganesha server config below #GLOBAL####################################################### TIMEOUTidle = 600 renegotiation = no FIPS = no options = NO_SSLv2 options = NO_SSLv3 options = SINGLE_DH_USE options = SINGLE_ECDH_USE options = CIPHER_SERVER_PREFERENCE syslog = yes debug = 7 setuid = nobody setgid = nobody chroot = /var/empty/stunnel libwrap = yes service = MC-nfsd ; cd /var/empty; mkdir -p stunnel/etc; cd stunnel/etc; ; echo 'MC-nfsd: ALL EXCEPT 5.6.7.8' >> hosts.deny; ; chcon -t stunnel_etc_t hosts.deny curve = secp521r1 #CREDENTIALS################################################## verify = 4 CAfile = /etc/stunnel/nfs-tls.pem cert = /etc/stunnel/nfs-tls.pem #ROLE######################################################### connect = 127.0.0.1:2049 I have had a look through the documentation and I believe I need to set transparent = source on the client side, and then set some ip tables firewall rules. Does anyone have a guide, or some advice on how to get this to work. Generally what happens if I set the firewall rules, on the client, and set the transparent to source I just get connection closed by remote host. I never actually see the traffic leave the client host. To confirm, when not using transparent everything works correctly, accept the server side sees the connection coming from 127.0.0.1 Thanks Rgds Steve. The future has already arrived. It's just not evenly distributed yet - William Gibson Steven Relf - Technical Authority: Cloud Native Infrastructure srelf(a)ukcloud.com +44 1252 936019 / +44 7500 085 864 www.ukcloud.com A8, Cody Technology Park, Ively Road, Farnborough, GU14 0LX Notice: This message contains information that may be privileged or confidential and is the property of UKCloud Ltd. It is intended only for the person to whom it is addressed. If you are not the intended recipient, you are not authorised to read, print, retain, copy, disseminate, distribute, or use this message or any part thereof. If you receive this message in error, please notify the sender immediately and delete all copies of this message. UKCloud reserves the right to monitor all e-mail communications through its networks. UKCloud Ltd is registered in England and Wales: Company No: 07619797. Registered office: Hartham Park, Hartham, Corsham, Wiltshire SN13 0RP.

1 0

Used Multi Service
by Ramin Malekgahsemi 09 Jul '20

09 Jul '20

Hi Dear Friends I used Stunnel 5.49 32 bit For server This Config cert = C:\Program Files\stunnel\config\stunnel.pem fips = no [socks] accept = 443 connect = 1010 If define New Service For example cert = C:\Program Files\stunnel\config\stunnel.pem fips = no [s1] accept = 443 connect = 1010 [s2] accept = 445 connect = 1020 Service Not work How Solve This Problem THX

1 0

Re: [stunnel-users] winsock permissions issue (i.e. WSAEACCES 10013 / 0x271D)
by flash＠vicsmba.com 29 Jun '20

29 Jun '20

It will never be entirely random. Some range more than likely. You may be able to restrict the range. You could certainly set stunnel up for each “random” port. E VICS, LLC Eric S Eberhard 2933 W Middle Verde Rd Camp Verde, AZ 86322 928-567-3727 (land line) 928-301-7537 (cell phone) http://www.vicsmba.com <http://www.vicsmba.com/> https://www.facebook.com/groups/286143052248115 From: stunnel-users <stunnel-users-bounces(a)stunnel.org> On Behalf Of Brent Kimberley Sent: Friday, June 26, 2020 4:11 PM To: Thomas Ward <teward(a)thomas-ward.net>; stunnel-users(a)stunnel.org; Coviello, Paul <pcoviello(a)ccsusa.com> Subject: Re: [stunnel-users] winsock permissions issue (i.e. WSAEACCES 10013 / 0x271D) What are you trying to accomplish? I.e. Is TLS the best wrapper for the job? (i.e. TCP over TCP.) On Thursday, June 25, 2020, 11:44:38 a.m. EDT, Coviello, Paul <pcoviello(a)ccsusa.com <mailto:pcoviello@ccsusa.com> > wrote: Maybe, I also just found out that the port it goes out is not what I thought it was. It’s a random port. Not sure if it’s doable now. Thanks for your help

1 0

winsock permissions issue (i.e. WSAEACCES 10013 / 0x271D)
by Brent Kimberley 26 Jun '20

26 Jun '20

Hi Paul. With respect to your winsock permissions issue (i.e. WSAEACCES 10013 / 0x271D), did you open the port in your firewall and authorize communication via anti-virus? > Hi I have a windows 10 machine that I?ve got this running on, and the > screen shot below is what I keep getting and I am unable to find the issue, > > Can someone please shed some light as to what is wrong? And what else you may need.

3 4

server is down message
by Coviello, Paul 24 Jun '20

24 Jun '20

Hi I have a windows 10 machine that I've got this running on, and the screen shot below is what I keep getting and I am unable to find the issue, Can someone please shed some light as to what is wrong? And what else you may need. Thanks Paul [cid:image001.png@01D6495F.3B2D17A0]

3 4

Fwd: Update Stunnel 5.44 - Ubuntu 18.04
by Juan Amores 19 Jun '20

19 Jun '20

---------- Forwarded message --------- De: Juan Amores <canalamores(a)gmail.com> Date: jue., 18 jun. 2020 a las 20:03 Subject: *Update Stunnel 5.44 - Ubuntu 18.04* I installed Stunnel4 via apt-get install command. The version you installed for me is old (5.44), I have tried updating it via apt-get update but there are no updated packages in Ubuntu 18.04. Is there a way to update it and keep it working as a service like before? Attached screenshot with linux, stunnel and openssl versions. Any other additional data you need to provide just ask me. [image: stunnel.png] Greetings, Juan Amores

2 2

Fw: How to configure STunnel so that new log file is automatically created after a certain threshold
by Muhammad Muneeb Hashmi 17 Jun '20

17 Jun '20

Hi STunnel Team, Is there a way where a new STunnel log file is automatically generated when the current log file reaches a certain threshold in Windows. Any help will be highly appreciated. Regards,Muneeb

1 0

CAPI_GET_KEY:cryptacquirecontext error
by Michael S. Chusovitin 04 Jun '20

04 Jun '20

Dear stunnel users, whereas Stunnel itself looks fine for me, it seems to have some troubles with CAPI engine. The one I cannot manage still is *"CAPI_GET_KEY:cryptacquirecontext error"* logged when attempting to export the client's key. I tried to add *"engineCtrl = csp_name:Microsoft Enhanced RSA and AES Cryptographic Provider"* to stunnel.conf, and then *"CAPI_**CTX_SET_PROVNAME:**cryptacquirecontext error"* was logged instead of the above one. The same certificate is used happily with no-capi configuration. I suppose the problem is somehow related to sha256 signature in the certificate. Are there any chances to workaround? Stunnel version is 5.48 with OpenSSL 1.0.2o-fips. (in this very case I need to use 32bit version, so no possibility to upgrade). Thanks!

2 5

Re: [stunnel-users] Need help connecting Eudora to mail servers
by Phil Smith III 28 May '20

28 May '20

I'm still tinkering here, trying to get Yahoo+stunnel working to help out Joe Slaven. I still can't get it to work. Yahoo userid is y_phsiii. I manually logged into Yahoo and verified that I have the right password. The config section: -- [Yahoo] client = yes accept = localhost:1234 connect = pop.mail.yahoo.com:995 sslVersion = TLSv1.2 -- Eudora has -- POPPort=1234 -- for that personality. Log: -- 2020.05.27 11:36:48 LOG5[29]: Service [Yahoo] accepted connection from 127.0.0.1:50287 2020.05.27 11:36:48 LOG6[29]: failover: priority, starting at entry #0 2020.05.27 11:36:48 LOG6[29]: s_connect: connecting 98.137.157.46:995 2020.05.27 11:36:48 LOG5[29]: s_connect: connected 98.137.157.46:995 2020.05.27 11:36:48 LOG5[29]: Service [Yahoo] connected remote server from 192.168.1.4:50288 2020.05.27 11:36:48 LOG6[29]: SNI: sending servername: pop.mail.yahoo.com 2020.05.27 11:36:48 LOG6[29]: Peer certificate not required 2020.05.27 11:36:48 LOG6[29]: Certificate verification disabled 2020.05.27 11:36:48 LOG6[29]: Certificate verification disabled 2020.05.27 11:36:48 LOG6[29]: Client certificate not requested 2020.05.27 11:36:48 LOG6[29]: Session id: F26CEF81143B96B8BD0D2C504CBB534082242BF337D3D29DE911CD5AC05B6DE0 2020.05.27 11:36:48 LOG6[29]: TLS connected: new session negotiated 2020.05.27 11:36:48 LOG6[29]: TLSv1.2 ciphersuite: ECDHE-RSA-AES256-GCM-SHA384 (256-bit encryption) 2020.05.27 11:37:48 LOG6[29]: TLS socket closed (SSL_read) 2020.05.27 11:37:48 LOG5[29]: Connection closed: 38 byte(s) sent to TLS, 130 byte(s) sent to socket -- Eudora error (Personality is named just "y" since it's just a test): -- y, Logging into POP Server, PASS [11:33:29 AM] There has been an error transferring your mail. I said: PASS <shhhh! Don't tell anyone.> and then the POP server (y_phsiii(a)127.0.0.1) said: ERR [SYS/TEMP] Server error - Please try again later. --

3 3

blackhole problem - please help to tune stunnel
by Michael S. Chusovitin 28 May '20

28 May '20

Dear stunnel users, please advise how to solve the following: - an Application connects to stunnel-client (installed at the same machine); - stunnel-client tries to connect to stunnel-server (remote), fails and sends RST to the App; - but the App has already sent some datagrams to stunnel-client during TIMEOUTconnect period and they aren't transferred to stunnel-server. Is there any way to make stunnel-client delay its ACK to the App until the connection to stunnel-server is established? Thanks!

1 0

Need help connecting Eudora to mail servers
by Joe Slaven 26 May '20

26 May '20

Below are the contents of my config file: Question 1. Is the output statement correct? I cannot find the log file. I expected it to be in the stunnel folder in the Program (x86) folders Question 2. Is it permissable to install code for all three email servers, using the same ports and settings in Eudora? Question 3. All three have error messages in Eudora ie none of them work via stunnel. Can anyone please show me where I am going wrong? I have listed the error message for Gmail below. There is no log file that I can find. Bigpond error message: There has been an error transferring your mail POP server [JoeS4814(a)127.0.0.1 ] said ERR[AUTH] Username and Password not accepted thanks for helping out, Joe ;*************************************************************** ; STUNNEL LOG output = stunnel.log ; ;***************************************************************** ;******************************************************* ;Yahoo connection ;******************************************************* [yahoo-pop3] client = yes accept = 127.0.0.1:110 connect = pop.mail.yahoo.com:995 verify = 2 CAfile = ca-certs.pem checkHost = pop.mail.yahoo.com OCSPaia = yes [yahoo-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.mail.yahoo.com:465 verify = 2 CAfile = ca-certs.pem checkHost = smtp.mail.yahoo.com OCSPaia = yes ; ;******************************************************* ;Bigpond connection ;***************************************************** [bigpond-pop3] client = yes accept = 127.0.0.1:110 connect = pop.bigpond.com:995 verify = 2 CAfile = ca-certs.pem checkHost = pop.bigpond.com OCSPaia = yes [bigpond-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.bigpond.com:465 verify = 2 CAfile = ca-certs.pem checkHost = smtp.bigpond.com OCSPaia = yes ;************************************* ;Gmail connection ;******************************************** [gmail-pop3] client = yes accept = 127.0.0.1:110 connect = pop.gmail.com:995 verifyChain = yes CAfile = ca-certs.pem checkHost = pop.gmail.com OCSPaia = yes [gmail-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.gmail.com:465 verifyChain = yes CAfile = ca-certs.pem checkHost = smtp.gmail.com OCSPaia = yes

3 7

[PATCH] Use libcap to keep CAP_NET_RAW so that IP_TRANSPARENT works after setuid
by Simon Arlott 26 May '20

26 May '20

Use libcap to keep CAP_NET_RAW so that IP_TRANSPARENT works after setuid I'd like to be able to use "transparent = source" and "setuid" to drop full root privileges. It is possible to do this by keeping either the CAP_NET_ADMIN or CAP_NET_RAW capability. This is based on https://github.com/dlundquist/sniproxy/commit/0ab83f8f45f73ba18ea5f84ccccd9… The capability will not be inherited if "exec" is used: Name: stunnel Uid: 996 996 996 996 Gid: 988 988 988 988 ... CapInh: 0000000000000000 CapPrm: 0000000000002000 CapEff: 0000000000002000 --- It would be possible to keep CAP_NET_BIND_SERVICE too, to allow reloading a configuration that binds to ports below 1024. I don't think stunnel should keep this capability unless a service needs it at startup. CREDITS.md | 25 ++++++++++ configure.ac | 33 ++++++++++++++ doc/stunnel.8.in | 2 +- doc/stunnel.html.in | 2 +- doc/stunnel.pod.in | 2 +- src/stunnel.c | 109 ++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 170 insertions(+), 3 deletions(-) diff --git a/CREDITS.md b/CREDITS.md index 6567392..5f12f92 100644 --- a/CREDITS.md +++ b/CREDITS.md @@ -27,6 +27,31 @@ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. +libcap usage in version 5.57 (from sniproxy): +Copyright 2017 Dustin Lundquist <dustin(a)null-ptr.net> +Copyright 2020 Simon Arlott + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions +are met: +1. Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. +2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE HOLDERS OR +CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + Several bugfixes and improvements mostly in versions 3.xx: * Brian Hatch <bri(a)stunnel.org> diff --git a/configure.ac b/configure.ac index 9c2b2b6..f454710 100644 --- a/configure.ac +++ b/configure.ac @@ -352,6 +352,39 @@ AC_ARG_ENABLE(systemd, ] ) +# Disable POSIX capabilities support +AC_MSG_CHECKING([whether to enable POSIX capabilities support]) +AC_ARG_ENABLE(libcap, +[ --disable-libcap disable POSIX capabilities support], + [ + case "$enableval" in + yes) AC_MSG_RESULT([yes]) + AC_SEARCH_LIBS([cap_get_proc], [cap]) + AC_DEFINE([USE_LIBCAP], [1], + [Define to 1 to enable POSIX capabilities]) + ;; + no) AC_MSG_RESULT([no]) + ;; + *) AC_MSG_RESULT([error]) + AC_MSG_ERROR([Bad value \"${enableval}\"]) + ;; + esac + ], + [ + AC_MSG_RESULT([autodetecting]) + AC_SEARCH_LIBS([cap_get_proc], [cap], + [ AC_CHECK_HEADERS([sys/capability.h], [ + AC_DEFINE([USE_LIBCAP], [1], + [Define to 1 to enable POSIX capabilities]) + AC_MSG_NOTICE([libcap support enabled]) + ], [ + AC_MSG_NOTICE([libcap header not found]) + ]) ], [ + AC_MSG_NOTICE([libcap library not found]) + ]) + ] +) + # Disable use of libwrap (TCP wrappers) # it should be the last check! AC_MSG_CHECKING([whether to enable TCP wrappers support]) diff --git a/doc/stunnel.8.in b/doc/stunnel.8.in index b3df016..4b121bc 100644 --- a/doc/stunnel.8.in +++ b/doc/stunnel.8.in @@ -1251,7 +1251,7 @@ setuid .RS 4 .Sp The use of the 'setuid' option will also prevent \fBstunnel\fR from binding to privileged -(<1024) ports during configuration reloading. +(<1024) ports during configuration reloading or enabling of 'transparent = source'. .Sp When the 'chroot' option is used, \fBstunnel\fR will look for all its files (including the configuration file, certificates, the log file and the pid file) within the chroot diff --git a/doc/stunnel.html.in b/doc/stunnel.html.in index ee9e5d9..95a1b51 100644 --- a/doc/stunnel.html.in +++ b/doc/stunnel.html.in @@ -1506,7 +1506,7 @@ </li> </ul> -The use of the 'setuid' option will also prevent stunnel from binding to privileged (<1024) ports during configuration reloading. +The use of the 'setuid' option will also prevent stunnel from binding to privileged (<1024) ports during configuration reloading or enabling of 'transparent = source'. When the 'chroot' option is used, stunnel will look for all its files (including the configuration file, certificates, the log file and the pid file) within the chroot jail. diff --git a/doc/stunnel.pod.in b/doc/stunnel.pod.in index e9abd4f..8f14fd9 100644 --- a/doc/stunnel.pod.in +++ b/doc/stunnel.pod.in @@ -1360,7 +1360,7 @@ setuid =back The use of the 'setuid' option will also prevent B<stunnel> from binding to privileged -(<1024) ports during configuration reloading. +(<1024) ports during configuration reloading or enabling of 'transparent = source'. When the 'chroot' option is used, B<stunnel> will look for all its files (including the configuration file, certificates, the log file and the pid file) within the chroot diff --git a/src/stunnel.c b/src/stunnel.c index 8121e0c..b84ce72 100644 --- a/src/stunnel.c +++ b/src/stunnel.c @@ -58,6 +58,11 @@ #endif /* USE_WIN32 */ +#ifdef USE_LIBCAP +#include <sys/prctl.h> +#include <sys/capability.h> +#endif + /**************************************** prototypes */ #ifdef __INNOTEK_LIBC__ @@ -191,6 +196,59 @@ int drop_privileges(int critical) { #ifdef HAVE_SETGROUPS gid_t gr_list[1]; #endif +#ifdef USE_LIBCAP + SERVICE_OPTIONS *opt; + int need_caps = 0; + cap_t caps; + cap_value_t cap_list[1]; + int keepcaps = 0; +#endif + +#ifdef USE_LIBCAP + for (opt = service_options.next; opt; opt = opt->next) { + if (opt->option.transparent_src) { /* transparent_src is enabled for this service */ + need_caps = 1; + break; + } + } + + /* add permitted flag to capability needed for IP_TRANSPARENT */ + if (CAP_IS_SUPPORTED(CAP_NET_RAW) && need_caps) { + const char *cap_failed = NULL; + + caps = cap_get_proc(); + if (caps != NULL) { + cap_list[0] = CAP_NET_RAW; + if (cap_set_flag(caps, CAP_PERMITTED, 1, cap_list, CAP_SET) == -1) { + cap_failed = "cap_set_flags[before]"; + goto cap_cleanup_before; + } + + if (cap_set_proc(caps) == -1) { + cap_failed = "cap_set_proc[before]"; + goto cap_cleanup_before; + } + + /* keep capabilities after setuid */ + if (prctl(PR_SET_KEEPCAPS, 1, 0, 0, 0) == -1) { + cap_failed = "prctl[before]"; + goto cap_cleanup_before; + } else { + keepcaps = 1; + } + +cap_cleanup_before: + cap_free(caps); + if (cap_failed != NULL && critical) { + sockerror(cap_failed); + return 1; + } + } else if (critical) { + sockerror("cap_get_proc[before]"); + return 1; + } + } +#endif /* set uid and gid */ if(service_options.gid) { @@ -212,6 +270,57 @@ int drop_privileges(int critical) { return 1; } } + +#ifdef USE_LIBCAP + /* enable capability needed for IP_TRANSPARENT */ + if (CAP_IS_SUPPORTED(CAP_NET_RAW) && need_caps) { + const char *cap_failed = NULL; + + caps = cap_get_proc(); + if (caps != NULL) { + /* remove every capability from the list */ + cap_clear(caps); + cap_list[0] = CAP_NET_RAW; + + /* take back capability */ + if (cap_set_flag(caps, CAP_PERMITTED, 1, cap_list, CAP_SET) == -1) { + cap_failed = "cap_set_flag[after]"; + goto cap_cleanup_after; + } + + if (cap_set_flag(caps, CAP_EFFECTIVE, 1, cap_list, CAP_SET) == -1) { + cap_failed = "cap_set_flag[after]"; + goto cap_cleanup_after; + } + + if (cap_set_proc(caps) == -1) { + cap_failed = "cap_set_proc[after]"; + goto cap_cleanup_after; + } + /* from now on it the new capability list is active, + * no other capabilities may be enabled */ + +cap_cleanup_after: + cap_free(caps); + if (cap_failed != NULL && critical) { + sockerror(cap_failed); + return 1; + } + } else if (critical) { + sockerror("cap_get_proc[after]"); + return 1; + } + } + + /* disable keeping capabilities across setuid */ + if (keepcaps && prctl(PR_SET_KEEPCAPS, 0, 0, 0, 0) == -1) { + if (critical) { + sockerror("prctl"); + return 1; + } + } +#endif + #endif /* standard Unix */ return 0; } -- 2.17.1 -- Simon Arlott

1 0

Re: [stunnel-users] stunnel-users Digest, Vol 190, Issue 13
by Brent Kimberley 24 May '20

24 May '20

Try: * Verifying the openssl version(s) * Verifying the 192.168.1.96:51400 port (nmap, netstat, etc.) * Verifying connection (openssl s_client -connect 192.168.1.96:51400) * Verifying the ciphersuite * Verifying the certificate >>Is there a configuration item in stunnel to at least tell it to chill out a little and not try reconnecting 1000 times a second? Not sure. TIMEOUTclose? On Sunday, May 24, 2020, 6:00:07 a.m. EDT, stunnel-users-request(a)stunnel.org <stunnel-users-request(a)stunnel.org> wrote: Send stunnel-users mailing list submissions to stunnel-users(a)stunnel.org To subscribe or unsubscribe via the World Wide Web, visit https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users or, via email, send a message with subject or body 'help' to stunnel-users-request(a)stunnel.org You can reach the person managing the list at stunnel-users-owner(a)stunnel.org When replying, please edit your Subject line so it is more specific than "Re: Contents of stunnel-users digest..." Today's Topics: 1. /var/log/syslog is filling up with stunnel errors (digitek) 2. Re: /var/log/syslog is filling up with stunnel errors (Thomas Ward) ---------------------------------------------------------------------- Message: 1 Date: Sat, 23 May 2020 18:17:49 -0500 From: digitek <digitek(a)charter.net> To: stunnel-users(a)stunnel.org Subject: [stunnel-users] /var/log/syslog is filling up with stunnel errors Message-ID: <434e3153-d94c-bb71-fc51-1716db588290(a)charter.net> Content-Type: text/plain; charset=windows-1252; format=flowed Hi, I've been running stunnel for years to shuttle my syslogs off to a syslog server. It's performed flawlessly up until a few weeks ago. Recently, rebooting my syslog server results in the clients filling up /var/log/syslog with messages like below[1]. I'm not certain where the issue is. We've gone through some Ubuntu 16.04 > 18.04 upgrades on some hosts, as well as the syslog server. Is there a configuration item in stunnel to at least tell it to chill out a little and not try reconnecting 1000 times a second? My config from a client is here[2] Last time this happened (few weeks ago) I googled around and found the TIMEOUTconnect parameter to try and get stunnel to at least wait 10 seconds before attempting another connect, but guess it doesn't work that way. Any thoughts? The stunnel client is on Ubuntu 18.04. I'd rather not compile out-of-band for the latest stunnel version unless I must. I"m assuming this is a config issue on my end. The version I'm stuck with right now is stunnel 5.44 on x86_64-pc-linux-gnu platform Thanks! [1] May 23 17:42:04 shuriken stunnel: LOG5[44922183]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socket May 23 17:42:04 shuriken stunnel: LOG5[44922184]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55440 May 23 17:42:04 shuriken stunnel: LOG3[44922184]: s_connect: connect 192.168.1.96:51400: Connection refused (111) May 23 17:42:04 shuriken stunnel: LOG3[44922184]: No more addresses to connect May 23 17:42:04 shuriken stunnel: LOG5[44922184]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socket May 23 17:42:04 shuriken stunnel: LOG5[44922185]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55444 May 23 17:42:04 shuriken stunnel: LOG3[44922185]: s_connect: connect 192.168.1.96:51400: Connection refused (111) May 23 17:42:04 shuriken stunnel: LOG3[44922185]: No more addresses to connect [2] client = yes cert = /etc/stunnel/shared/stunnel.pem pid = /var/run/stunnel4/syslog_stunnel.pid [syslog_tunnel] accept = 127.0.0.1:5140 connect = 192.168.1.96:51400 TIMEOUTconnect = 10 ------------------------------ Message: 2 Date: Sat, 23 May 2020 20:09:48 -0400 From: Thomas Ward <teward(a)thomas-ward.net> To: digitek <digitek(a)charter.net>, stunnel-users(a)stunnel.org Subject: Re: [stunnel-users] /var/log/syslog is filling up with stunnel errors Message-ID: <49V0wH16GWz2QGD(a)mail.syn-ack.link> Content-Type: text/plain; charset="utf-8" This is indicative of the remote server not running on the right ports normally or actively blocking you, given the "Connection Refused" errors.? Verify your system stunnel is on actually can connect to the specified IP and port combo independently of stunnel to start with.Sent from my Sprint Samsung Galaxy Note10+. -------- Original message --------From: digitek <digitek(a)charter.net> Date: 5/23/20 19:18 (GMT-05:00) To: stunnel-users(a)stunnel.org Subject: [stunnel-users] /var/log/syslog is filling up with stunnel errors Hi,I've been running stunnel for years to shuttle my syslogs off to a syslog server. It's performed flawlessly up until a few weeks ago. Recently, rebooting my syslog server results in the clients filling up /var/log/syslog with messages like below[1].I'm not certain where the issue is. We've gone through some Ubuntu 16.04 > 18.04 upgrades on some hosts, as well as the syslog server. Is there a configuration item in stunnel to at least tell it to chill out a little and not try reconnecting 1000 times a second?? My config from a client is here[2]Last time this happened (few weeks ago) I googled around and found the TIMEOUTconnect parameter to try and get stunnel to at least wait 10 seconds before attempting another connect, but guess it doesn't work that way.Any thoughts? The stu nnel client is on Ubuntu 18.04. I'd rather not compile out-of-band for the latest stunnel version unless I must. I"m assuming this is a config issue on my end.? The version I'm stuck with right now is stunnel 5.44 on x86_64-pc-linux-gnu platformThanks![1]May 23 17:42:04 shuriken stunnel: LOG5[44922183]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socketMay 23 17:42:04 shuriken stunnel: LOG5[44922184]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55440May 23 17:42:04 shuriken stunnel: LOG3[44922184]: s_connect: connect 192.168.1.96:51400: Connection refused (111)May 23 17:42:04 shuriken stunnel: LOG3[44922184]: No more addresses to connectMay 23 17:42:04 shuriken stunnel: LOG5[44922184]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socketMay 23 17:42:04 shuriken stunnel: LOG5[44922185]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55444May 23 17:42:04 shuriken stunnel: LOG3[44922185]: s_connect: connect 192.168.1.96:51400: Conn ection refused (111)May 23 17:42:04 shuriken stunnel: LOG3[44922185]: No more addresses to connect[2]client?? = yescert???? = /etc/stunnel/shared/stunnel.pempid????? = /var/run/stunnel4/syslog_stunnel.pid[syslog_tunnel]accept?? = 127.0.0.1:5140connect? = 192.168.1.96:51400TIMEOUTconnect = 10_______________________________________________stunnel-users mailing liststunnel-users@stunnel.orghttps://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users

1 0

/var/log/syslog is filling up with stunnel errors
by digitek 24 May '20

24 May '20

Hi, I've been running stunnel for years to shuttle my syslogs off to a syslog server. It's performed flawlessly up until a few weeks ago. Recently, rebooting my syslog server results in the clients filling up /var/log/syslog with messages like below[1]. I'm not certain where the issue is. We've gone through some Ubuntu 16.04 > 18.04 upgrades on some hosts, as well as the syslog server. Is there a configuration item in stunnel to at least tell it to chill out a little and not try reconnecting 1000 times a second? My config from a client is here[2] Last time this happened (few weeks ago) I googled around and found the TIMEOUTconnect parameter to try and get stunnel to at least wait 10 seconds before attempting another connect, but guess it doesn't work that way. Any thoughts? The stunnel client is on Ubuntu 18.04. I'd rather not compile out-of-band for the latest stunnel version unless I must. I"m assuming this is a config issue on my end. The version I'm stuck with right now is stunnel 5.44 on x86_64-pc-linux-gnu platform Thanks! [1] May 23 17:42:04 shuriken stunnel: LOG5[44922183]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socket May 23 17:42:04 shuriken stunnel: LOG5[44922184]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55440 May 23 17:42:04 shuriken stunnel: LOG3[44922184]: s_connect: connect 192.168.1.96:51400: Connection refused (111) May 23 17:42:04 shuriken stunnel: LOG3[44922184]: No more addresses to connect May 23 17:42:04 shuriken stunnel: LOG5[44922184]: Connection reset: 0 byte(s) sent to TLS, 0 byte(s) sent to socket May 23 17:42:04 shuriken stunnel: LOG5[44922185]: Service [syslog_tunnel] accepted connection from 127.0.0.1:55444 May 23 17:42:04 shuriken stunnel: LOG3[44922185]: s_connect: connect 192.168.1.96:51400: Connection refused (111) May 23 17:42:04 shuriken stunnel: LOG3[44922185]: No more addresses to connect [2] client = yes cert = /etc/stunnel/shared/stunnel.pem pid = /var/run/stunnel4/syslog_stunnel.pid [syslog_tunnel] accept = 127.0.0.1:5140 connect = 192.168.1.96:51400 TIMEOUTconnect = 10

2 3

Re: [stunnel-users] Eudora -> Yahoo problems
by phs3stuff＠cox.net 20 May '20

20 May '20

Have you looked at the stunnel log? FWIW, I use stunnel to cox.net with Eudora (well, my wife does--I get to make it work) and just have the protocol, accept, and connect statements and it works. You might try commenting those out and see, but the log is probably more interesting. -----Original Message----- From: stunnel-users <stunnel-users-bounces(a)stunnel.org> On Behalf Of Joe Slaven Sent: Tuesday, May 19, 2020 8:48 PM To: stunnel-users(a)stunnel.org Subject: [stunnel-users] Eudora -> Yahoo problems Hello, new to stunnel, my first attempt is connecting Eudora to Yahoo Mail. My Yahoo Mail password and username are not being recognised when it is trying to access server 127.0.0.1 Can anyone suggest why this will not work? regards, Joe ******************************************************************************************************** ERROR MESSAGE There has been an error transferring your mail. POP Server [joeslaven(a)127.0.0.1] said ERR [AUTH] Username and password not accepted STUNNEL CONFIG FILE SETTINGS [yahoo-pop3] client = yes accept = 127.0.0.1:110 connect = pop.mail.yahoo.com:995 verify = 2 CAfile = ca-certs.pem checkHost = pop.mail.yahoo.com OCSPaia = yes [yahoo-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.mail.yahoo.com:465 verify = 2 CAfile = ca-certs.pem checkHost = smtp.mail.yahoo.com OCSPaia = yes EUDORA SETTINGS Checking Mail->Mail POP Server: 127.0.0.1 Sending Mail->SMTP Server: 127.0.0.1 Use submission port (587) is UNchecked Secure Sockets when Sending or Receiving : Never Default port settings : SMPT Port (25) POP Port (110) _______________________________________________ stunnel-users mailing list stunnel-users(a)stunnel.org https://www.stunnel.org/cgi-bin/mailman/listinfo/stunnel-users

1 0

Eudora -> Yahoo problems
by Joe Slaven 20 May '20

20 May '20

Hello, new to stunnel, my first attempt is connecting Eudora to Yahoo Mail. My Yahoo Mail password and username are not being recognised when it is trying to access server 127.0.0.1 Can anyone suggest why this will not work? regards, Joe ******************************************************************************************************** ERROR MESSAGE There has been an error transferring your mail. POP Server [joeslaven(a)127.0.0.1] said ERR [AUTH] Username and password not accepted STUNNEL CONFIG FILE SETTINGS [yahoo-pop3] client = yes accept = 127.0.0.1:110 connect = pop.mail.yahoo.com:995 verify = 2 CAfile = ca-certs.pem checkHost = pop.mail.yahoo.com OCSPaia = yes [yahoo-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.mail.yahoo.com:465 verify = 2 CAfile = ca-certs.pem checkHost = smtp.mail.yahoo.com OCSPaia = yes EUDORA SETTINGS Checking Mail->Mail POP Server: 127.0.0.1 Sending Mail->SMTP Server: 127.0.0.1 Use submission port (587) is UNchecked Secure Sockets when Sending or Receiving : Never Default port settings : SMPT Port (25) POP Port (110)

1 0

Eudora -> Yahoo problems
by Joe Slaven 20 May '20

20 May '20

Hello, new to stunnel, my first attempt is connecting Eudora to Yahoo Mail. My Yahoo Mail password and username are not being recognised when it is trying to access server 127.0.0.1 Can anyone suggest why this will not work? regards, Joe ******************************************************************************************************** ERROR MESSAGE There has been an error transferring your mail. POP Server [joeslaven(a)127.0.0.1] said ERR [AUTH] Username and password not accepted STUNNEL CONFIG FILE SETTINGS [yahoo-pop3] client = yes accept = 127.0.0.1:110 connect = pop.mail.yahoo.com:995 verify = 2 CAfile = ca-certs.pem checkHost = pop.mail.yahoo.com OCSPaia = yes [yahoo-smtp] client = yes accept = 127.0.0.1:25 connect = smtp.mail.yahoo.com:465 verify = 2 CAfile = ca-certs.pem checkHost = smtp.mail.yahoo.com OCSPaia = yes EUDORA SETTINGS Checking Mail->Mail POP Server: 127.0.0.1 Sending Mail->SMTP Server: 127.0.0.1 Use submission port (587) is UNchecked Secure Sockets when Sending or Receiving : Never Default port settings : SMPT Port (25) POP Port (110)

1 0

Eudora to Yahoo Mail -amended
by Joe Slaven 19 May '20

19 May '20

Hello, totally new to stunnel. Can anyone please point me in the direction of example config code to connect Eudora to POP Yahoo mail? Also what do I put in Eudora ? many thanks, Joe

1 0

Lost password
by Joe Slaven 19 May '20

19 May '20

Hello, I forgot to record my password used to subscribe to the list. Can anyone please tell me how I can reset it? Many thanks, joe

1 0

Eudora to Yahoo Mail
by Joe Slaven 19 May '20

19 May '20

Hello, totally new to stunnel. Can anyone please point me in the direction of example config code to connect Eudora to POP Yahoo mail? many thanks, Joe

1 0

newer version for VMS
by Coviello, Paul 19 May '20

19 May '20

Has anyone put together a package for OpenVMS particularly 8.4-1h1 from VSI? And if not is there a way to use an updated version of SSL in the current version I have? Which I believe is 4.20 FACS1>stunnel -version stunnel undefined on vax-OpenVMS with OpenSSL 0.9.8h 28 May 2008 Threading:PTHREAD Sockets:POLL,IPv4 Global options debug = 5 pid = SYS$LOGIN:STUNNEL.PID RNDbytes = 64 RNDoverwrite = yes Service-level options cert = stunnel.pem ciphers = AES:ALL:!aNULL:!eNULL:+RC4:@STRENGTH key = stunnel.pem session = 300 seconds sslVersion = SSLv3 for client, all for server TIMEOUTbusy = 300 seconds TIMEOUTclose = 60 seconds TIMEOUTconnect = 10 seconds TIMEOUTidle = 43200 seconds verify = none

1 0

Windows tray icon
by Phil Smith III 19 May '20

19 May '20

If you hover on the stunnel tray icon, it just says "Server is idle". Right-clicking shows various options; none of those say "stunnel". Yes, they can go to "About", but it would be nice if the hover said "Stunnel server is idle" (and, perhaps, the "Exit" option said "Exit stunnel"). My user (my wife!) keeps forgetting what it is, hovering, and then still asking me what it is. Just a WIBNI. stunnel has been a lifesaver: she's wedded to Eudora, which doesn't support modern TLS, so when our ISP went all-TLS we had a problem. I also use it: I have a Bayesian mail proxy (K9) that doesn't support TLS, so my connections go MUA==>K9==>stunnel==>ISP. Any chance of this happening next rev? ...phsiii

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

stunnel-users