[sr-dev] [SR-Users] Info: initialization of libssl 1.1.+ with PTHREAD_PROCESS_SHARED attribute for pthread mutex and rwlock
Daniel-Constantin Mierla
miconda at gmail.com
Wed Sep 18 17:26:35 CEST 2019
You have to install the debug symbols package, otherwise the backtrace
doesn't show any useful information. It just require to install that
package and then get again the backtraces. The package name should be
kamailio-dbg.
Cheers,
Daniel
On 18.09.19 16:37, Andrew Chen wrote:
> Sorry..forgot bt full output
>
> (gdb) bt full
>
> #0 0x0000562fcc3544c0 in ?? ()
>
> No symbol table info available.
>
> #1 0x0000562fcc354632 in ?? ()
>
> No symbol table info available.
>
> #2 0x0000562fcc354974 in local_timer_run ()
>
> No symbol table info available.
>
> #3 0x0000562fcc3fb256 in ?? ()
>
> No symbol table info available.
>
> #4 0x0000562fcc3fb64c in tcp_receive_loop ()
>
> No symbol table info available.
>
> #5 0x0000562fcc2acf52 in tcp_init_children ()
>
> No symbol table info available.
>
> #6 0x0000562fcc1ad94a in main_loop ()
>
> No symbol table info available.
>
> #7 0x0000562fcc1b4a7b in main ()
>
> No symbol table info available.
>
> (gdb)
>
>
> On Wed, Sep 18, 2019 at 10:35 AM Andrew Chen <achen at fuze.com
> <mailto:achen at fuze.com>> wrote:
>
> backtrace output:
>
>
> (gdb) bt
>
> #0 0x0000562fcc3544c0 in ?? ()
>
> #1 0x0000562fcc354632 in ?? ()
>
> #2 0x0000562fcc354974 in local_timer_run ()
>
> #3 0x0000562fcc3fb256 in ?? ()
>
> #4 0x0000562fcc3fb64c in tcp_receive_loop ()
>
> #5 0x0000562fcc2acf52 in tcp_init_children ()
>
> #6 0x0000562fcc1ad94a in main_loop ()
>
> #7 0x0000562fcc1b4a7b in main ()
>
> (gdb)
>
>
>
>
> On Wed, Sep 18, 2019 at 10:01 AM Andrew Chen <achen at fuze.com
> <mailto:achen at fuze.com>> wrote:
>
> I'll look for the core.
>
> As part of our release process, I don't think we can
> release master branch of Kamailio.
>
> On Wed, Sep 18, 2019 at 9:54 AM Daniel-Constantin Mierla
> <miconda at gmail.com <mailto:miconda at gmail.com>> wrote:
>
> Hello,
>
> did you get core files? If yes, it would be good to see
> the gdb bracktraces.
>
> Have you used the pre-loaded library workaround?
>
> The code was not backported to branch 5.2 yet, I wanted to
> wait a bit and see if there are reports with good/bad
> results. If you want to use it, then you have to install
> nightly builds from master branch (development version,
> should be 5.3.0-pre1 at this moment).
>
> Cheers,
> Daniel
>
> On 18.09.19 15:26, Andrew Chen wrote:
>> Hi Daniel,
>> Which apt repository can we get this workaround? Today
>> I noticed our Kamailio 5.2.4 cored during one of our load
>> test sessions:
>>
>> Wed Sep 18 09:31:14 2019] kamailio[26997]: segfault at
>> 7f8501366000 ip 00007f8501e58585 sp 00007fffd5c84248
>> error 4 in libcrypto.so.1.1[7f8501dd4000+29b000]
>>
>> [Wed Sep 18 10:12:29 2019] kamailio[27110]: segfault at
>> 7f1a8eea1000 ip 00007f1a8f993585 sp 00007ffddc8848c8
>> error 4 in libcrypto.so.1.1[7f1a8f90f000+29b000]
>>
>> [Wed Sep 18 10:12:43 2019] kamailio[11060]: segfault at
>> 20 ip 00007f0a85dac128 sp 00007fff6847aad0 error 4 in
>> libcrypto.so.1.1[7f0a85c40000+29b000]
>>
>> [Wed Sep 18 10:12:44 2019] kamailio[11059]: segfault at
>> f0 ip 00007f0a85cc43ea sp 00007fff6847a9e8 error 4 in
>> libcrypto.so.1.1[7f0a85c40000+29b000]
>>
>> [Wed Sep 18 10:13:02 2019] kamailio[11187]: segfault at
>> 7f75e06c8000 ip 00007f75e11ba585 sp 00007ffe29b47c88
>> error 4 in libcrypto.so.1.1[7f75e1136000+29b000]
>>
>> [Wed Sep 18 10:13:17 2019] kamailio[11445]: segfault at
>> 7f0f3743f000 ip 00007f0f37f31585 sp 00007ffde54daa38
>> error 4 in libcrypto.so.1.1[7f0f37ead000+29b000]
>>
>> [Wed Sep 18 10:13:36 2019] kamailio[11644]: segfault at
>> 7f804309c000 ip 00007f8043b8e585 sp 00007ffe89fe8228
>> error 4 in libcrypto.so.1.1[7f8043b0a000+29b000]
>>
>> [Wed Sep 18 10:13:52 2019] kamailio[11799]: segfault at
>> 7ff59c7e2000 ip 00007ff59d2d4585 sp 00007ffd69d76bb8
>> error 4 in libcrypto.so.1.1[7ff59d250000+29b000]
>>
>> [Wed Sep 18 10:13:52 2019] kamailio[11792]: segfault at
>> 7ff59c7e2000 ip 00007ff59d2d4585 sp 00007ffd69d76bb8
>> error 4 in libcrypto.so.1.1[7ff59d250000+29b000]
>>
>> [Wed Sep 18 12:48:01 2019] kamailio[12066]: segfault at
>> 7f22dca6d000 ip 00007f22dd55f585 sp 00007fffbc766ab8
>> error 4 in libcrypto.so.1.1[7f22dd4db000+29b000]
>>
>> [Wed Sep 18 12:48:14 2019] kamailio[16344]: segfault at
>> 7f7cca42d000 ip 00007f7ccaf1f585 sp 00007ffe99f16108
>> error 4 in libcrypto.so.1.1[7f7ccae9b000+29b000]
>>
>> [Wed Sep 18 12:48:32 2019] kamailio[16708]: segfault at
>> 7fb309141000 ip 00007fb309c33585 sp 00007ffd07a5f3c8
>> error 4 in libcrypto.so.1.1[7fb309baf000+29b000]
>>
>> [Wed Sep 18 12:48:49 2019] kamailio[16824]: segfault at
>> f0 ip 00007fc59095b3ea sp 00007ffe58a490b8 error 4 in
>> libcrypto.so.1.1[7fc5908d7000+29b000]
>>
>> [Wed Sep 18 12:49:04 2019] kamailio[16956]: segfault at
>> 7fcc22e58000 ip 00007fcc2394a585 sp 00007fffa9c52258
>> error 4 in libcrypto.so.1.1[7fcc238c6000+29b000]
>>
>> [Wed Sep 18 12:49:04 2019] kamailio[16945]: segfault at
>> f0 ip 00007fcc2394a3ea sp 00007fffa9c52128 error 4 in
>> libcrypto.so.1.1[7fcc238c6000+29b000]
>>
>> [Wed Sep 18 12:49:16 2019] kamailio[17244]: segfault at
>> 7f586ed20000 ip 00007f586f812585 sp 00007ffeae1adcd8
>> error 4 in libcrypto.so.1.1[7f586f78e000+29b000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17431]: segfault at 8
>> ip 0000562fcc44c5d5 sp 00007ffe825c9dd0 error 4 in
>> kamailio[562fcc17c000+4a2000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17386]: segfault at
>> 10 ip 00007f96c3735c3e sp 00007ffe825ca270 error 4 in
>> dialog.so[7f96c36d6000+be000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17423]: segfault at 0
>> ip 0000562fcc3544c0 sp 00007ffe825ca1f0 error 6 in
>> kamailio[562fcc17c000+4a2000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17378]: segfault at 0
>> ip 00007f96c5dd099f sp 00007ffe825c98e0 error 4 in
>> tm.so[7f96c5da5000+13f000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17417]: segfault at 0
>> ip 0000562fcc3544c0 sp 00007ffe825ca1f0 error 6 in
>> kamailio[562fcc17c000+4a2000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17416]: segfault at 0
>> ip 0000562fcc3544c0 sp 00007ffe825ca1f0 error 6 in
>> kamailio[562fcc17c000+4a2000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17380]: segfault at 0
>> ip 00007f96c5dd099f sp 00007ffe825c98e0 error 4 in
>> tm.so[7f96c5da5000+13f000]
>>
>> [Wed Sep 18 12:49:33 2019] kamailio[17381]: segfault at 0
>> ip 00007f96c5dd099f sp 00007ffe825c98e0 error 4 in
>> tm.so[7f96c5da5000+13f000]
>>
>> [Wed Sep 18 12:49:34 2019] kamailio[17387]: segfault at 8
>> ip 0000562fcc25d1f3 sp 00007ffe825ca370 error 6 in
>> kamailio[562fcc17c000+4a2000]
>>
>>
>> and I was wondering if this core is related.
>>
>>
>> Thanks
>>
>>
>> On Fri, Sep 13, 2019 at 5:18 PM Alex Balashov
>> <abalashov at evaristesys.com
>> <mailto:abalashov at evaristesys.com>> wrote:
>>
>> Hi Daniel,
>>
>> Thanks a lot for this work; it's definitely important
>> that Kamailio
>> build against OpenSSL v1.1, and I'll be happy to
>> contribute some testing
>> in a live environment in the coming weeks. Much
>> appreciated.
>>
>> Have a good weekend,
>>
>> -- Alex
>>
>> On Fri, Sep 13, 2019 at 01:34:12PM +0200,
>> Daniel-Constantin Mierla wrote:
>>
>> > Hello,
>> > after couple of months waiting for feedback about
>> the workaround with the
>> > preloaded shared library to set the
>> PTHREAD_PROCESS_SHARED attribute for
>> > pthread mutex/rwlock in order to avoid deadlocks
>> and as it seems to be the
>> > solution, I just pushed a few commits to move that
>> code in the core of
>> > kamailio.
>> >
>> > It this way the initialization with the
>> PTHREAD_PROCESS_SHARED attribute is
>> > done by default when libssl v1.1 is detected at
>> compile time, without the
>> > need of using the approach with a preloaded library.
>> >
>> > The shared library is still kept in the tls module:
>> > *
>> >
>> https://github.com/kamailio/kamailio/tree/master/src/modules/tls/utils/openssl_mutex_shared
>> >
>> > Probably is going to be removed in the near future.
>> >
>> > I am asking the community members that use Kamailio
>> on systems with libssl
>> > v1.1+ to test the master branch and give feedback.
>> The master branch is
>> > anyhow in testing phase for preparing the next
>> major release v5.3.0, so it
>> > will help for that as well.
>> >
>> > If no issue is going to be reported soon, the
>> commits will be backported to
>> > stable branches, to re-enable smooth use of
>> Kamailio with libssl v1.1+.
>> >
>> > When libssl v1.1+ is detected, the core links with
>> pthread, but that was
>> > already present in the Makefiles.defs for different
>> compile options. The
>> > pthread more or less a default library on
>> linuxes/unixes, so this should
>> > not introduce any major dependency to the Kamailio
>> core.
>> >
>> > Cheers,
>> > Daniel
>> >
>> > --
>> > Daniel-Constantin Mierla - http://www.asipto.com
>> > http://twitter.com/#!/miconda -
>> http://www.linkedin.com/in/miconda
>>
>> > _______________________________________________
>> > Kamailio (SER) - Users Mailing List
>> > sr-users at lists.kamailio.org
>> <mailto:sr-users at lists.kamailio.org>
>> >
>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>
>>
>> --
>> Alex Balashov | Principal | Evariste Systems LLC
>>
>> Tel: +1-706-510-6800 / +1-800-250-5920 (toll-free)
>> Web: http://www.evaristesys.com/,
>> http://www.csrpswitch.com/
>>
>> _______________________________________________
>> Kamailio (SER) - Users Mailing List
>> sr-users at lists.kamailio.org
>> <mailto:sr-users at lists.kamailio.org>
>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>>
>>
>>
>> --
>> Andy Chen
>> Sr. Telephony Lead Engineer
>> 415 516 5535 (M)
>> achen@ <mailto:achen at thinkingphones.com>fuze.com
>> <http://fuze.com>
>>
>>
>> *Confidentiality Notice: The information contained in
>> this e-mail and any
>> attachments may be confidential. If you are not an
>> intended recipient, you
>> are hereby notified that any dissemination, distribution
>> or copying of this
>> e-mail is strictly prohibited. If you have received this
>> e-mail in error,
>> please notify the sender and permanently delete the
>> e-mail and any
>> attachments immediately. You should not retain, copy or
>> use this e-mail or
>> any attachment for any purpose, nor disclose all or any
>> part of the
>> contents to any other person. Thank you.*
>>
>> _______________________________________________
>> Kamailio (SER) - Users Mailing List
>> sr-users at lists.kamailio.org <mailto:sr-users at lists.kamailio.org>
>> https://lists.kamailio.org/cgi-bin/mailman/listinfo/sr-users
>
> --
> Daniel-Constantin Mierla -- www.asipto.com <http://www.asipto.com>
> www.twitter.com/miconda <http://www.twitter.com/miconda> -- www.linkedin.com/in/miconda <http://www.linkedin.com/in/miconda>
> Kamailio Advanced Training, Oct 21-23, 2019, Berlin, Germany -- https://asipto.com/u/kat
>
>
>
> --
> Andy Chen
> Sr. Telephony Lead Engineer
> 415 516 5535 (M)
> achen@ <mailto:achen at thinkingphones.com>fuze.com <http://fuze.com>
>
>
>
> --
> Andy Chen
> Sr. Telephony Lead Engineer
> 415 516 5535 (M)
> achen@ <mailto:achen at thinkingphones.com>fuze.com <http://fuze.com>
>
>
>
> --
> Andy Chen
> Sr. Telephony Lead Engineer
> 415 516 5535 (M)
> achen@ <mailto:achen at thinkingphones.com>fuze.com <http://fuze.com>
>
>
> *Confidentiality Notice: The information contained in this e-mail and any
> attachments may be confidential. If you are not an intended recipient, you
> are hereby notified that any dissemination, distribution or copying of
> this
> e-mail is strictly prohibited. If you have received this e-mail in error,
> please notify the sender and permanently delete the e-mail and any
> attachments immediately. You should not retain, copy or use this e-mail or
> any attachment for any purpose, nor disclose all or any part of the
> contents to any other person. Thank you.*
--
Daniel-Constantin Mierla -- www.asipto.com
www.twitter.com/miconda -- www.linkedin.com/in/miconda
Kamailio Advanced Training, Oct 21-23, 2019, Berlin, Germany -- https://asipto.com/u/kat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.kamailio.org/pipermail/sr-dev/attachments/20190918/5d5f2c74/attachment-0001.html>
More information about the sr-dev
mailing list