Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
Hi Anca,
On 20.04.2011, at 13:24, Anca Vamanu wrote:
> Hi John,
>
> Thanks for the logs.
> Do you get the memory statistics from opensips statistics or from top?
> In the logs I don't see a shared memory leak..
Hm... Maybe the memory leak in the system libraries?
Memory usage is calculated as follows:
# ps aux | head -n 1
USER PID %CPU %MEMVSZ RSS TTY STAT START TIME COMMAND
# ps aux | grep -E "/usr/sbin/opensips -[P]"
opensips 8389 0.0 0.0 451444 9880 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8395 0.2 0.0 451448 5216 ?SApr19 2:06
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8396 0.0 0.0 451448 5184 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8398 0.0 0.0 451452 5160 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8399 0.0 0.1 451684 12732 ?SApr19 0:10
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8400 0.0 0.1 451704 12708 ?SApr19 0:09
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8402 0.0 0.1 451708 12644 ?SApr19 0:09
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8403 0.0 0.1 451708 12640 ?SApr19 0:10
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8405 0.0 0.0 451448 5088 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8406 0.0 0.0 451448 5088 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8409 0.0 0.0 451448 5088 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8410 0.0 0.0 451448 5088 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8411 0.0 0.0 450644 3148 ?SApr19 0:07
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8412 0.0 0.1 451448 11572 ?SApr19 0:11
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8413 0.0 0.0 451444 5304 ?SApr19 0:05
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8414 0.0 0.0 451452 9660 ?SApr19 0:01
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8418 0.0 0.0 451452 9336 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8426 0.0 0.0 451452 9684 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8428 0.0 0.0 451452 5920 ?SApr19 0:00
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
opensips 8429 0.0 0.0 451416 6544 ?SApr19 0:01
/usr/sbin/opensips -P /var/run/opensips/opensips.pid -m 256 -u opensips -g
opensips
# ps aux | grep -E "/usr/sbin/opensips -[P]" | awk '{a += $6} END {print a}'
157684
i.e. sum of RSS value, where RSS (man ps)
RSS - resident set size, the non-swapped physical memory that a task has used
(in kiloBytes). (alias rssize, rsz).
How can I help?
> On 04/20/2011 10:18 AM, John Khvatov wrote:
>> Hello Anca,
>>
>> On 19.04.2011, at 13:21, Anca Vamanu wrote:
>>
>>> Hi John,
>>>
>>> That looks very bad...
>>> Please help me investigate this.
>>>
>>> What you can do is to compile with memory debugging support - in
>>> Makefile.defs search -DDBG_QM_MALLOC line, uncomment it, move it above
>>> -DF_MALLOC line and comment this line with -DF_MALLOC. Then recompile and
>>> reinstall all. Also set mem_dump=1 in your config.
>> I've built OpenSIPS from latest SVN head (branch 1.6) as you said.
>>
>>> After a couple of hours, when you see that the memory is high - you can
>>> either shutdown opensips or ask it to dump the memory by calling 'kill
>>> -SIGUSR1 lowest_pid' , where lowest_pid is the smallest pid when doing 'ps
>>> aux | grep opensips'. It will dump lots of info in the log file(this why is
>>> better to do this quite soon - maybe after 1-2 hours in the rhythm the
>>> memory increases for you). In the logs you will see al
Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
Hello Anca, On 19.04.2011, at 13:21, Anca Vamanu wrote: > Hi John, > > That looks very bad... > Please help me investigate this. > > What you can do is to compile with memory debugging support - in > Makefile.defs search -DDBG_QM_MALLOC line, uncomment it, move it above > -DF_MALLOC line and comment this line with -DF_MALLOC. Then recompile and > reinstall all. Also set mem_dump=1 in your config. I've built OpenSIPS from latest SVN head (branch 1.6) as you said. > After a couple of hours, when you see that the memory is high - you can > either shutdown opensips or ask it to dump the memory by calling 'kill > -SIGUSR1 lowest_pid' , where lowest_pid is the smallest pid when doing 'ps > aux | grep opensips'. It will dump lots of info in the log file(this why is > better to do this quite soon - maybe after 1-2 hours in the rhythm the memory > increases for you). In the logs you will see all the chunks of memory and > where they were allocated. Look from the end if there is a function repeated > many times. You can also send the part of the log with the dump to me ( not > on the list, because it will be quite large). Done. SIGUSR1: http://dev.sgu.ru/pub/memdump.log.gz SIGTERM: http://dev.sgu.ru/pub/memdump.log.shutdown.gz > On 04/19/2011 10:30 AM, John Khvatov wrote: >> Hello Anca, >> >> Sorry, the memory leak is still here: >> >> http://dev.sgu.ru/pub/opensips_memory_usage.2.png >> >> BTW, dialog module was not used in my script. I've removed loadmodule before >> update. >> >> >> On 12.04.2011, at 0:00, John Khvatov wrote: >> >>> Hello Anca, >>> >>> Thank you for the advice. I built OpenSIPS from rev7863. BTW, rev7824 was >>> used in production. >>> >>> I'll let you know the result within a few days. >>> >>> >>> On 11.04.2011, at 17:48, Anca Vamanu wrote: >>> >>>> Hi John, >>>> >>>> Can you please take the pua and pua_usrloc and dialog modules from svn? >>>> There were some problems discovered in there and fixed. >>>> The svn version of these modules ( 1.6 branch) is stable, we are using it >>>> in production. >>>> >>>> Regards, >>>> -- >>>> Anca Vamanu >>>> OpenSIPS Developer >>>> >>>> >>>> >>>> On 04/11/2011 12:45 AM, John Khvatov wrote: >>>>> On 10.04.2011, at 8:35, Ronald Cepres wrote: >>>>> >>>>>> Have you tried increasing the amount of allocated memory for OpenSIPS? >>>>> I have same problem with OpenSIPS since 1.6.0 release. OpenSIPS has huge >>>>> memory leak somewhere. Increasing the amount of allocated memory helps >>>>> only delay crash. >>>>> >>>>> Memory usage graph (1.6.4, KkB is MB): >>>>> http://dev.sgu.ru/pub/opensips_memory_usage.png >>>>> >>>>> I use the following modules: >>>>> >>>>> $ grep loadmodule opensips.cfg >>>>> loadmodule "signaling.so" >>>>> loadmodule "sl.so" >>>>> loadmodule "tm.so" >>>>> loadmodule "maxfwd.so" >>>>> loadmodule "rr.so" >>>>> loadmodule "textops.so" >>>>> loadmodule "localcache.so" >>>>> loadmodule "mi_fifo.so" >>>>> loadmodule "mi_datagram.so" >>>>> loadmodule "db_flatstore.so" >>>>> loadmodule "siptrace.so" >>>>> loadmodule "db_text.so" >>>>> loadmodule "db_postgres.so" >>>>> loadmodule "db_http.so" >>>>> loadmodule "usrloc.so" >>>>> loadmodule "registrar.so" >>>>> loadmodule "auth.so" >>>>> loadmodule "auth_db.so" >>>>> loadmodule "uri.so" >>>>> loadmodule "domain.so" >>>>> loadmodule "dialog.so" >>>>> loadmodule "mediaproxy.so" >>>>> loadmodule "nat_traversal.so" >>>>> loadmodule "permissions.so" >>>>> loadmodule "alias_db.so" >>>>> loadmodule "avpops.so" >>>>> loadmodule "mi_xmlrpc.so" >>>>> loadmodule "dialplan.so" >>>>> loadmodule "pike.so" >>>>> loadmodule "pua.so" >>>>> loadmodule "pua_usrloc.so" >>>>> >>>>> -- >>>>> WBR, John Khvatov >>> -- >>> WBR, John Khvatov > > > ___ > Users mailing list > Users@lists.opensips.org > http://lists.opensips.org/cgi-bin/mailman/listinfo/users -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
Hi, Eric. On 19.04.2011, at 12:16, Erik Dekkers wrote: > Offtopic, but with what tool did you create this graph? Looks very nice It's zabbix monitoring system. > -Oorspronkelijk bericht- > Van: users-boun...@lists.opensips.org > [mailto:users-boun...@lists.opensips.org] Namens John Khvatov > > Hello Anca, > > Sorry, the memory leak is still here: > > http://dev.sgu.ru/pub/opensips_memory_usage.2.png > > BTW, dialog module was not used in my script. I've removed loadmodule before > update. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
Hello Anca, Sorry, the memory leak is still here: http://dev.sgu.ru/pub/opensips_memory_usage.2.png BTW, dialog module was not used in my script. I've removed loadmodule before update. On 12.04.2011, at 0:00, John Khvatov wrote: > Hello Anca, > > Thank you for the advice. I built OpenSIPS from rev7863. BTW, rev7824 was > used in production. > > I'll let you know the result within a few days. > > > On 11.04.2011, at 17:48, Anca Vamanu wrote: > >> Hi John, >> >> Can you please take the pua and pua_usrloc and dialog modules from svn? >> There were some problems discovered in there and fixed. >> The svn version of these modules ( 1.6 branch) is stable, we are using it in >> production. >> >> Regards, >> -- >> Anca Vamanu >> OpenSIPS Developer >> >> >> >> On 04/11/2011 12:45 AM, John Khvatov wrote: >>> >>> On 10.04.2011, at 8:35, Ronald Cepres wrote: >>> >>>> Have you tried increasing the amount of allocated memory for OpenSIPS? >>> >>> I have same problem with OpenSIPS since 1.6.0 release. OpenSIPS has huge >>> memory leak somewhere. Increasing the amount of allocated memory helps only >>> delay crash. >>> >>> Memory usage graph (1.6.4, KkB is MB): >>> http://dev.sgu.ru/pub/opensips_memory_usage.png >>> >>> I use the following modules: >>> >>> $ grep loadmodule opensips.cfg >>> loadmodule "signaling.so" >>> loadmodule "sl.so" >>> loadmodule "tm.so" >>> loadmodule "maxfwd.so" >>> loadmodule "rr.so" >>> loadmodule "textops.so" >>> loadmodule "localcache.so" >>> loadmodule "mi_fifo.so" >>> loadmodule "mi_datagram.so" >>> loadmodule "db_flatstore.so" >>> loadmodule "siptrace.so" >>> loadmodule "db_text.so" >>> loadmodule "db_postgres.so" >>> loadmodule "db_http.so" >>> loadmodule "usrloc.so" >>> loadmodule "registrar.so" >>> loadmodule "auth.so" >>> loadmodule "auth_db.so" >>> loadmodule "uri.so" >>> loadmodule "domain.so" >>> loadmodule "dialog.so" >>> loadmodule "mediaproxy.so" >>> loadmodule "nat_traversal.so" >>> loadmodule "permissions.so" >>> loadmodule "alias_db.so" >>> loadmodule "avpops.so" >>> loadmodule "mi_xmlrpc.so" >>> loadmodule "dialplan.so" >>> loadmodule "pike.so" >>> loadmodule "pua.so" >>> loadmodule "pua_usrloc.so" >>> >>> -- >>> WBR, John Khvatov >> ___ >> Users mailing list >> Users@lists.opensips.org >> http://lists.opensips.org/cgi-bin/mailman/listinfo/users > > -- > WBR, John Khvatov > -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
Hello Anca, Thank you for the advice. I built OpenSIPS from rev7863. BTW, rev7824 was used in production. I'll let you know the result within a few days. On 11.04.2011, at 17:48, Anca Vamanu wrote: > Hi John, > > Can you please take the pua and pua_usrloc and dialog modules from svn? There > were some problems discovered in there and fixed. > The svn version of these modules ( 1.6 branch) is stable, we are using it in > production. > > Regards, > -- > Anca Vamanu > OpenSIPS Developer > > > On 04/11/2011 12:45 AM, John Khvatov wrote: >> >> >> On 10.04.2011, at 8:35, Ronald Cepres wrote: >> >>> Have you tried increasing the amount of allocated memory for OpenSIPS? >> >> I have same problem with OpenSIPS since 1.6.0 release. OpenSIPS has huge >> memory leak somewhere. Increasing the amount of allocated memory helps only >> delay crash. >> >> Memory usage graph (1.6.4, KkB is MB): >> http://dev.sgu.ru/pub/opensips_memory_usage.png >> >> I use the following modules: >> >> $ grep loadmodule opensips.cfg >> loadmodule "signaling.so" >> loadmodule "sl.so" >> loadmodule "tm.so" >> loadmodule "maxfwd.so" >> loadmodule "rr.so" >> loadmodule "textops.so" >> loadmodule "localcache.so" >> loadmodule "mi_fifo.so" >> loadmodule "mi_datagram.so" >> loadmodule "db_flatstore.so" >> loadmodule "siptrace.so" >> loadmodule "db_text.so" >> loadmodule "db_postgres.so" >> loadmodule "db_http.so" >> loadmodule "usrloc.so" >> loadmodule "registrar.so" >> loadmodule "auth.so" >> loadmodule "auth_db.so" >> loadmodule "uri.so" >> loadmodule "domain.so" >> loadmodule "dialog.so" >> loadmodule "mediaproxy.so" >> loadmodule "nat_traversal.so" >> loadmodule "permissions.so" >> loadmodule "alias_db.so" >> loadmodule "avpops.so" >> loadmodule "mi_xmlrpc.so" >> loadmodule "dialplan.so" >> loadmodule "pike.so" >> loadmodule "pua.so" >> loadmodule "pua_usrloc.so" >> >> -- >> WBR, John Khvatov > ___ > Users mailing list > Users@lists.opensips.org > http://lists.opensips.org/cgi-bin/mailman/listinfo/users -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Opensips crashing due to out of memory error
On 10.04.2011, at 8:35, Ronald Cepres wrote: > Have you tried increasing the amount of allocated memory for OpenSIPS? > > On Sun, Apr 10, 2011 at 5:53 AM, Nauman Sulaiman > wrote: > Hi, I'm getting random crashes of Opensips 1.6.2, here are the entries in the > log, seems to be out of memory, how should i try to solve this issue. > > ERROR:tm:_reply_light: failed to allocate shmem buffer > ERROR:tm:_reply_light: failed to allocate shmem buffer > ERROR:tm:relay_reply: no more share memory > ERROR:core:new_avp: no more shm mem > ERROR:core:add_avp: Failed to create new avp structure I have same problem with OpenSIPS since 1.6.0 release. OpenSIPS has huge memory leak somewhere. Increasing the amount of allocated memory helps only delay crash. Memory usage graph (1.6.4, KkB is MB): http://dev.sgu.ru/pub/opensips_memory_usage.png I use the following modules: $ grep loadmodule opensips.cfg loadmodule "signaling.so" loadmodule "sl.so" loadmodule "tm.so" loadmodule "maxfwd.so" loadmodule "rr.so" loadmodule "textops.so" loadmodule "localcache.so" loadmodule "mi_fifo.so" loadmodule "mi_datagram.so" loadmodule "db_flatstore.so" loadmodule "siptrace.so" loadmodule "db_text.so" loadmodule "db_postgres.so" loadmodule "db_http.so" loadmodule "usrloc.so" loadmodule "registrar.so" loadmodule "auth.so" loadmodule "auth_db.so" loadmodule "uri.so" loadmodule "domain.so" loadmodule "dialog.so" loadmodule "mediaproxy.so" loadmodule "nat_traversal.so" loadmodule "permissions.so" loadmodule "alias_db.so" loadmodule "avpops.so" loadmodule "mi_xmlrpc.so" loadmodule "dialplan.so" loadmodule "pike.so" loadmodule "pua.so" loadmodule "pua_usrloc.so" -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] media-relay exception
Hello. On 29.03.2011, at 14:14, Saúl Ibarra Corretgé wrote: > On 29/3/11 11:42 AM, n...@uni-petrol.com wrote: >> >> >> Because of media-relay running in openvz virtual container it inherit >> kernel modules like ip_table, ip_conntrack, nfnetlink from host node, so >> yesm ip_tables module like others loade before media-relay start. >> Below is a list of running modules. >> > > I haven't tried MediaProxy within a OpenVZ container, so I can't say how this > could affect MediaProxy, sorry. > >> Unfortunately it is very hard to migrate to python > 2.4, because CentOS >> < 6.0 >> don't support other versions and hard depend on python 2.4. >> >> Now I'm trying install Scientific Linux 6.0 in openvz virtual environment >> to test python 2.6. >> > > We develop, test and deploy MediaProxy on Debian based non virtualized > systems so chances are that we'll not be able to help you if you run into > issues we didn't experience because we don't use such scenarios. > > Many people are running MediaProxy on CentOS-like systems, but I can't recall > if anyone is using OpenVZ containers. You may want to test it in a standalone > machine first, and once you get it working you can put it in the container > and test again. I tried media-relay within a OpenVz container. As far as I can remember, I got same exceptions during call establishing: > mediaproxy.interfaces.system._conntrack.Error: Table does not exist (do you > need to insmod?)" OS: debian/ubuntu (2.6.26-2-openvz-amd64), python: 2.6. I think that this is openvz-related problem. I use media-relay on hardware machine now, but it would be nice to have ability deploy media-relay in virtual machine. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] ERROR:core:tcp_read_req: bad request
Hello all. I could be wrong but it looks like bug in parser: Mar 22 17:26:22 sip /usr/sbin/opensips[3005]: ERROR:core:tcp_read_req: bad request, state=4, error=4 buf:#012REGISTER sip:tipmeet.com SIP/2.0#015#012Via: SIP/2.0/TCP 192.168.1.100:58279;branch=z9hG4bK-d8754z-e162c0293960910c-1---d8754z-;rport#015#012Max-Forwards: 70#015#012Contact: #015#012To: "96210185"#015#012From: "96210185";tag=6a11c927#015#012Call-ID: YTJhZmEzN2EzM2NkNGU1ZWY0ZDU3MTZlZTM3Y2Q1MWQ.#015#012CSeq: 2 REGISTER#015#012Expires: 3600#015#012Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, NOTIFY, MESSAGE, SUBSCRIBE, INFO#015#012User-Agent: Bria release 2.5 RC4 stamp 47242#015#012Authorization: Digest username="tipmeet.com#012#01296210185",realm="tipmeet.com",nonce="4d88dbdb0b9759f36d9f0a8fbac1ae1a3aea9963bae6",uri="sip:tipmeet.com",response="bc7d38b1bb7b190d94c8ad6a4cd4599a",cnonce="3264a3294f724a3d1cd0ca276593b9ba",nc=0001,qop=auth,algorithm=MD5#015#012Content-Length: 0#015#012#015#012#012parsed:#012REGISTER sip:tipmeet.com SIP/2.0#015#012Via: SIP/2.0/TCP 192.168.1.100:58279;branch=z9hG4bK-d8754z-e162c0293960910c-1---d8754z-;rport#015#012Max-Forwards: 70#015#012Contact: #015#012To: "96210185"#015#012From: "96210185";tag=6a11c927#015#012Call-ID: YTJhZmEzN2EzM2NkNGU1ZWY0ZDU3MTZlZTM3Y2Q1MWQ.#015#012CSeq: 2 REGISTER#015#012Expires: 3600#015#012Allow: INVITE, ACK, CANCEL, OPTIONS, BYE, REFER, NOTIFY, MESSAGE, SUBSCRIBE, INFO#015#012User-Agent: Bria release 2.5 RC4 stamp 47242#015#012Authorization: Digest username="tipmeet.com#012#012 Pay attention to this header: Authorization: Digest username="tipmeet.com#012#01296210185" Quoted string with two NL chars: "tipmeet.com\n\n96210185" did not parsed correctly. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] opensips.pid on Red Hat
On 09.03.2011, at 21:34, Ovidiu Sas wrote: > Run `opensips -h' and check -P option. > That should help you out with setting the pid file. Yep. Toyima, as workaround, add "-P /var/run/opensips.pid" to OPTIONS var in /etc/sysconfig/opensips: [root@host ~]# cat /etc/sysconfig/opensips # Add extra options here OPTIONS="-P /var/run/opensips.pid" Then restart opensips service. Pid-file should be created in /var/run/opensips.pid... I'll fix fedora/epel init-script later to fix this issue. > On Wed, Mar 9, 2011 at 1:19 PM, Toyima Dias wrote: >> Hello, >> >> I can't find the opensips.pid on Red Hat, it is not in the path: >> /var/run/opensips.pid >> >> Has somebody installed OpenSIPS on Red Hat and be able to find the pid of >> the process opensips? i'm trying to monitor the opensips process with >> "monit" and i need the full path of the process id... >> >> Regards! >> >> ___ >> Users mailing list >> Users@lists.opensips.org >> http://lists.opensips.org/cgi-bin/mailman/listinfo/users >> >> > > _______ > Users mailing list > Users@lists.opensips.org > http://lists.opensips.org/cgi-bin/mailman/listinfo/users -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] PATCH: avp_db_query() return codes
Hello all.
I wrote a small patch for avpops module:
http://dev.sgu.ru/pub/0001-avp_db_query-improved-error-reporting.patch
Current implementation of avp_db_query returns -1 on internal error or empty
result. Thus, current function interface does not allow to write script with
storage error handling.
Patched avp_db_query() returns:
-3 on internal error
-1 on empty result
1 on success
With patched avp_db_query() we can write following logic in opensips.cfg:
avp_db_query("SELECT 1 WHERE false", "$avp(s:name)")
if ($retcode == -3) {
xlog("Error while querying database. Handle error.");
}
else if ($retcode == -1) {
xlog("Handle empty result.");
}
else {
xlog("Handle result");
}
I've tested this patch with 1.6.4.
Please, review my patch and apply to upstream.
--
WBR, John Khvatov
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Installing OpenSIPS on Red Hat
Hello, Adrian. Unfortunately, MediaProxy is not available from Fedora/EPEL repositories. This should be fixed :) On 02.02.2011, at 16:55, Adrian Georgescu wrote: > Anyone doing Fedora RPMs for MediaProxy? > > I have only a CentOS link here: > > http://mediaproxy-ng.org/wiki/InstallationGuide > > Adrian -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Installing OpenSIPS on Red Hat
Hi Bogdan. On 02.02.2011, at 15:45, Bogdan-Andrei Iancu wrote: > Hi John, > > Can I list you here: > http://www.opensips.org/Resources/Downloads#osipmi Sure. Also, you can update 'Fedora' (link: 'http://fedoraproject.org', latest version of OpenSIPS available in all currently supported Fedora branches) and add 'EPEL' with link: http://fedoraproject.org/wiki/EPEL. > John Khvatov wrote: >> Hello. >> >> I maintain OpenSIPS package in fedora/epel. >> >> Of course, let me know if you have issues with OpenSIPS rpm package from >> official fedora/epel repos. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] $auth.resp script variable
Hello.
There is problem with $auth.resp script variable.
Line from opensips.cfg:
xlog("L_INFO", "$auth.resp\n");
opensips -f opensips.cfg -c results:
Feb 2 16:33:12 [24453] NOTICE:core:main: config file ok, exiting...
Runtime error:
Feb 2 16:34:27 aki /usr/sbin/opensips[24485]: ERROR:core:pv_parse_spec: pvar
"auth.resp" not found
Feb 2 16:34:27 aki /usr/sbin/opensips[24485]: ERROR:core:pv_parse_spec: wrong
char [p/112] in [$auth.resp#012] at [9 (0)]
Feb 2 16:34:27 aki /usr/sbin/opensips[24485]: ERROR:core:fix_actions: wrong
fomat [$auth.resp#012] for value param
Feb 2 16:34:27 aki /usr/sbin/opensips[24485]: ERROR:core:fix_actions: fixing
failed (code=-5) at cfg line 216
Feb 2 16:34:27 aki /usr/sbin/opensips[24485]: ERROR:core:main: failed to fix
configuration with err code -5
--
WBR, John Khvatov
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Installing OpenSIPS on Red Hat
Hello. I maintain OpenSIPS package in fedora/epel. Of course, let me know if you have issues with OpenSIPS rpm package from official fedora/epel repos. On 31.01.2011, at 20:03, Adrian Georgescu wrote: > Would the maintainers of those packages please step forward so that people > who need support know exactly who to ask when in need? > > This would help everyone to better understand who is maintaing what so that > there is no fear uncertainty and doubts but only sure things. > > Adrian -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Makefile patches (documentation installation)
Hello all. First patch fixes missed modules documentation installation by replacing 'install-modules' by 'install-modules-all' in 'install' target. The target 'install-modules-all' runs 'install-modules' and 'install-modules-doc' targets. Second patch makes 'install-modules-doc' safer to prevent overwriting global README file. The patches attached. 0002-fixed-install-modules-doc.patch Description: Binary data 0001-replaced-install-modules-by-install-modules-all-in-i.patch Description: Binary data -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS core dumps
Hi Ryan, On 16.11.2010, at 18:17, thrillerbee wrote: > John, > > I appreciate the suggestion, but, unfortunately, I don't use t_newtran(). > What led you to remove that? Hm, sorry… I was removing some functions from my opensips.cfg, then running various stress tests (REGISTER, INVITE etc.) and watching resident memory usage graph. Starting with 6.11.2010 (12:00) opensips working without t_newtrans() and everything fine: http://dev.sgu.ru/pub/without_tnewtrans.png (kKb == MB) Btw, now I use localcache for handling sip-request retransmits. With localcache resident memory usage also increases, but OpenSIPS does not crash… > On Tue, Nov 16, 2010 at 9:08 AM, John Khvatov wrote: > Hello. > > On 10.11.2010, at 19:19, thrillerbee wrote: >> Well, I spoke too soon - it's not just an issue with the opensipsctl fifo >> calls - looks more like a memory leak. It crashed again today, but I did >> get some errors in the syslog this time right before the crash: >> Nov 10 15:42:32 core1 /usr/local/sbin/opensips[27044]: >> ERROR:db_flatstore:new_flat_id: no pkg memory left >> Nov 10 15:42:32 core1 kernel: [5508366.582447] opensips[27044]: segfault at >> 10 ip 7fa7ff74c21f sp 7fffdc101700 error 4 in >> db_flatstore.so[7fa7ff749000+5000] >> To be thorough, I've attached the backtrace & output from print commands >> (although they're the same as before). > > I have the same issue. OpenSIPS crashes at db_flatstore.so due to lack of > memory (according resident memory usage graph). Recently I removed the use of > the t_newtran() function and localstore module in opensips.cfg and crash did > not happens for 11 days. Earlier crash happens every two-tree days. > > Do you use t_newtran() in your config? > > -- > WBR, John Khvatov > > > ___ > Users mailing list > Users@lists.opensips.org > http://lists.opensips.org/cgi-bin/mailman/listinfo/users > > > _______ > Users mailing list > Users@lists.opensips.org > http://lists.opensips.org/cgi-bin/mailman/listinfo/users -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Presence server performance
Hi Anca, On 18.11.2010, at 14:31, Anca Vamanu wrote: > Hi John, > > On 11/15/2010 05:34 PM, John Khvatov wrote: >> Hello all. >> >> I'm testing performance of OpenSIPS presence server. I got bad results... >> Any ideas how to improve it? >> [skiped] > Yes, this is not a good result .. the reason is that presence processing is > quite costly - it requires querying the database all the time - because the > Publish information is not stored in memory (the bodies might be very large). > So probably the conclusion of your tests is that if you want more then 100 > cps, you should use a cluster of presence servers :) . Сlearly. Thanks. But what about cluster of presence servers? Architecture with shared database and presence servers with enabled fallback2db option? In this case OpenSIPS for each request will do query to database because local cache may be expired. As result we have overloaded database and slow presence servers.. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] SIP presence and aliases
Hello all. Is it possible to implement sip presence with sip aliases? In case of PUBLISH it's seems that I can change presentity URI. But what about SUBSCRIBE/NOTIFY methods? -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] OpenSIPS core dumps
Hello. On 10.11.2010, at 19:19, thrillerbee wrote: > Well, I spoke too soon - it's not just an issue with the opensipsctl fifo > calls - looks more like a memory leak. It crashed again today, but I did get > some errors in the syslog this time right before the crash: > Nov 10 15:42:32 core1 /usr/local/sbin/opensips[27044]: > ERROR:db_flatstore:new_flat_id: no pkg memory left > Nov 10 15:42:32 core1 kernel: [5508366.582447] opensips[27044]: segfault at > 10 ip 7fa7ff74c21f sp 7fffdc101700 error 4 in > db_flatstore.so[7fa7ff749000+5000] > To be thorough, I've attached the backtrace & output from print commands > (although they're the same as before). I have the same issue. OpenSIPS crashes at db_flatstore.so due to lack of memory (according resident memory usage graph). Recently I removed the use of the t_newtran() function and localstore module in opensips.cfg and crash did not happens for 11 days. Earlier crash happens every two-tree days. Do you use t_newtran() in your config? -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Presence server performance
Hello. On 15.11.2010, at 18:34, John Khvatov wrote: > Hello all. > > I'm testing performance of OpenSIPS presence server. I got bad results... Any > ideas how to improve it? Has anyone done any performance tests with opensips-presence or using opensips-presence in production? Could you share any results? -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Presence server performance
Hello all. I'm testing performance of OpenSIPS presence server. I got bad results... Any ideas how to improve it? My test scenario (SIPp xml config: http://dev.sgu.ru/pub/pubsub.xml): SIPpPresence Server | PUBLISH | |>| | | | 200 OK | |<| | | | SUBSCRIBE (on publised uri) | |>| | | | 200 OK | |<| | | | NOTIFY | |<| | | | 200 OK | |>| After 10.000 calls (10.000 scenario) OpenSIPS began to write warnings about memory to log: WARNING:core:fm_malloc: Not enough free memory, will atempt defragmenation Results with 100 calls per second: Total calls: 74454 Calls per second: 16.485 cps Successful calls: 73772 Failed calls: 382 (With 10.000 "total calls" OpenSIPS shows 100 cps, almost zero failed calls) Results with 1000 calls per second: Total calls 20857 Calls per second: 12.696 cps Successful calls: 15360 Failed calls: 2497 SUBSCRIBE to one of 500.000 published user is running 1-2 sec.. System configuration. OS: Debian squeeze/sid (Not tuned) Processor: Intel Xeon E5345 RAM: 1G (OpenSIPS executed with -m 64 option. Memory is still available during tests) OpenSIPS: Version 1.6.3. Config is very simple: handle_subscribe() && handle_publish(). fallback2db disabled. Storage postgresql 8.4. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy: closing expired call when one party stops sending RTP
Hello! Thank you all for your replies! I'll try SST module to solve the problem. On Mon, 20 Sep 2010, Dan Pascu wrote: > > On 16 Sep 2010, at 14:52, John Khvatov wrote: > > > Hello all. > > > > I have a problem with RTP timeouts on media relay. > > > > Media relay sends "expired" command only when _both_ parties stop > > sending RTP. When we have one-directional RTP flow, media relay does > > not > > throws timeout. > > > > Is it a known behaviour? Any suggestions on how to fix it? > > This is not a bug, it's a feature. Consider the case where one party > uses voice activation and doesn't say anything for a few minutes, just > listens to the other party. Indeed, but FreeSWITCH will disconnect the call in this case. Hmm, actually that "doesn't say anything" is "doesn't send RTP"? -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy: closing expired call when one party stops sending RTP
Hello! On Thu, 16 Sep 2010, John Khvatov wrote: > Hello all. > > I have a problem with RTP timeouts on media relay. > > Media relay sends "expired" command only when _both_ parties stop > sending RTP. When we have one-directional RTP flow, media relay does not > throws timeout. I may be poorly explained the problem... We have a paid call UA1 <-> UA2. UA1 pays UA2 for the connection. RTP goes in both directions though a media relay. Imagine, UA1 crashes (softphone crashed or network problem) and does not send a BYE message. Now, the flow goes in one direction UA2 -> UA1. UA1 continues to pay the money (until UA2 has disconnected). Why media relay does not detect this call (connection) as expired? No support in the conntrack library? > Is it a known behaviour? Any suggestions on how to fix it? > > Thanks! > -- > WBR, John Khvatov -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Mediaproxy: closing expired call when one party stops sending RTP
Hello all. I have a problem with RTP timeouts on media relay. Media relay sends "expired" command only when _both_ parties stop sending RTP. When we have one-directional RTP flow, media relay does not throws timeout. Is it a known behaviour? Any suggestions on how to fix it? Thanks! -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Mediaproxy: media relay selection algorithm improvements
Hello all! We are working on building geo-distributed VoIP solution with Mediaproxy. I think, that the current media relay selection algorithm is not perfect and can be improved. In the current implementation we set IP address of a particular media relay that will be preferred by dispatcher. But what if the particular media relay will be unavailable? Media dispatcher will run normal selection algorithm and does not select neighbor of the particular media relay. I propose to implement the groups of media relays. Each media-relay can be registered on dispatcher with name and group. In the OpenSIPS config we can specify preferred media relay/relays by name or group. Consider the example. Several our relays in Frankfurt registered on dispatchers with group=frankfurt, in Moscow -- with group=moscow. When we set “$avp(s:media_relay_group) to "moscow", dispatcher will select one of media_relay hosted in Moscow. Also, we can simply add/remove media_relays in run-time. What do you think about this idea? Is it difficult to implement? -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] db_http module patch
Attached :)
On Wed, 08 Sep 2010, John Khvatov wrote:
> Hello all!
>
> I've attached patch, that fixes a form_result() function. This function
> does not work if a buffer (var_str buff) has a byte with high-order
> bit (in other words, a character with a code greater than 127).
>
> Please, review and fix this bug in upstream.
>
> --
> WBR, John Khvatov
--
WBR, John Khvatov
diff --git a/modules/db_http/db_http.so b/modules/db_http/db_http.so
index 4d845fc..4c79ac9 100755
Binary files a/modules/db_http/db_http.so and b/modules/db_http/db_http.so
differ
diff --git a/modules/db_http/http_dbase.c b/modules/db_http/http_dbase.c
index 4b20fb9..a3c31e5 100644
--- a/modules/db_http/http_dbase.c
+++ b/modules/db_http/http_dbase.c
@@ -474,7 +474,7 @@ int form_result(var_str buff, db_res_t** r)
while( cur < end )
{
- next = next_state[ state ][ (int)*cur ];
+ next = next_state[ state ][ (int)((unsigned char)*cur) ];
consume = 1;
if( state == OUT )
@@ -554,7 +554,7 @@ int form_result(var_str buff, db_res_t** r)
{
- next = next_state[ state ][ (int)*cur ];
+ next = next_state[ state ][ (int)((unsigned char)*cur) ];
consume = 1;
if( state == OUT )
{
diff --git a/modules/db_http/http_dbase.o b/modules/db_http/http_dbase.o
index b48f6c7..11213cc 100644
Binary files a/modules/db_http/http_dbase.o and b/modules/db_http/http_dbase.o
differ
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] db_http module patch
Hello all! I've attached patch, that fixes a form_result() function. This function does not work if a buffer (var_str buff) has a byte with high-order bit (in other words, a character with a code greater than 127). Please, review and fix this bug in upstream. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy closes ports
On Tue, 10 Aug 2010, Saúl Ibarra Corretgé wrote: > On 10/08/10 09:02, John Khvatov wrote: > > Hello Saúl, > > > > On Mon, 09 Aug 2010, Saúl Ibarra Corretgé wrote: > >> I'm not able to reproduce this. Could you please send me a SIP trace > >> captured with ngrep (strip sensible information or email it privately)? > >> Then I can check if something could make MediaProxy not work properly... > > > > Sorry, but problem is fixed in other component... > > > > I'm glad you got it solved, but can you give me some more detail on how > another component was triggering this behavior? It might be helpful for > others when they need to troubleshoot. It's very strange, but the problem solved after flush iptables rules on _remote_ machine, which send and receive media from mediaproxy (multi-line UA). Interesting, that after start firewall on remote machine again, we can't reproduce this issue. So, when I'll have more information, I'll report in this list of course. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy closes ports
Hello Saúl, On Mon, 09 Aug 2010, Saúl Ibarra Corretgé wrote: > I'm not able to reproduce this. Could you please send me a SIP trace > captured with ngrep (strip sensible information or email it privately)? > Then I can check if something could make MediaProxy not work properly... Sorry, but problem is fixed in other component... Saúl, thank you for your responses and help. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy closes ports
On Fri, Aug 6, 2010 at 6:37 PM, Saúl Ibarra Corretgé wrote: > > Yes, shouldn't be a problem. Any reINVITE taking place? What value do > you have for stream_timeout in MediaProxy configuration? No, this problem is reproduced in simple call (no reINVITEs). stream_timeout is not set in config.ini (default value is used). > > Also, can you paste the output of: > > cat /proc/sys/net/ipv4/netfilter/ip_conntrack_udp_timeout_stream 180 -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] Mediaproxy closes ports
On Fri, 06 Aug 2010, Saúl Ibarra Corretgé wrote: > Hi, > > On 06/08/10 15:56, John Khvatov wrote: > > Hello all! > > > > I have a problem with Mediaproxy: > > > > 1. Media session established (audio and video goes both directions). > > 2. After ~1 min media-relay closes audio or video ports. > > > > I've attached media-relay debug log to this email. > > > > Please, any ideas and suggestions how to resolve this problem. > > > > What do you see in the dispatcher logs? Is everything OK at the SIP level? Aug 6 09:15:33 sip media-dispatcher[22336]: debug: Issuing "update" command to relay at 193.169.160.2 Aug 6 09:16:33 sip media-dispatcher[22336]: last message repeated 2 times Aug 6 09:16:56 sip media-dispatcher[22336]: debug: Issuing "remove" command to relay at 193.169.160.2 "last message repeated 2 times" -- is it ok? About SIP. Nothing happened between beginning media session (09:15:39) and receiving BYE (09:16:56). Mediaproxy closes ports at 09:16:40. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] Mediaproxy closes ports
Hello all! I have a problem with Mediaproxy: 1. Media session established (audio and video goes both directions). 2. After ~1 min media-relay closes audio or video ports. I've attached media-relay debug log to this email. Please, any ideas and suggestions how to resolve this problem. Mediaproxy: 2.4.2. Linux: 2.6.26-2-openvz-amd64 -- WBR, John Khvatov Aug 6 09:15:33 earth media-relay[10532]: debug: Received new SDP offer Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53678 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53679 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53680 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53681 Aug 6 09:15:33 earth media-relay[10532]: debug: Added new stream: (audio) 193.169.160.50:8554 (RTP: Unknown, RTCP: Unknown) <-> 193.169.160.2:53678 <-> 193.16 9.160.2:53680 <-> Unknown (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53682 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53683 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53684 Aug 6 09:15:33 earth media-relay[10532]: mediaproxy.mediacontrol.StreamListenerProtocol starting on 53685 Aug 6 09:15:33 earth media-relay[10532]: debug: Added new stream: (video) 193.169.160.50:8458 (RTP: Unknown, RTCP: Unknown) <-> 193.169.160.2:53682 <-> 193.16 9.160.2:53684 <-> Unknown (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:33 earth media-relay[10532]: debug: created new session 1a632798455277b8: 93173...@tipmeet.com (ddd06077a660b1dc) --> di...@tipmeet.com Aug 6 09:15:33 earth media-relay[10532]: debug: updating existing session 1a632798455277b8: 93173...@tipmeet.com (ddd06077a660b1dc) --> di...@tipmeet.com Aug 6 09:15:33 earth media-relay[10532]: debug: Received updated SDP answer Aug 6 09:15:33 earth media-relay[10532]: debug: Unchanged stream: (audio) 193.169.160.50:8554 (RTP: Unknown, RTCP: Unknown) <-> 193.169.160.2:53678 <-> 193.16 9.160.2:53680 <-> Unknown (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:33 earth media-relay[10532]: debug: Unchanged stream: (video) 193.169.160.50:8458 (RTP: Unknown, RTCP: Unknown) <-> 193.169.160.2:53682 <-> 193.16 9.160.2:53684 <-> Unknown (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:39 earth media-relay[10532]: debug: updating existing session 1a632798455277b8: 93173...@tipmeet.com (ddd06077a660b1dc) --> di...@tipmeet.com Aug 6 09:15:39 earth media-relay[10532]: debug: Received updated SDP answer Aug 6 09:15:39 earth media-relay[10532]: debug: Got initial answer from callee for stream: (audio) 193.169.160.50:8554 (RTP: Unknown, RTCP: Unknown) <-> 193.1 69.160.2:53678 <-> 193.169.160.2:53680 <-> 193.169.160.50:9064 (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:39 earth media-relay[10532]: debug: Got initial answer from callee for stream: (video) 193.169.160.50:8458 (RTP: Unknown, RTCP: Unknown) <-> 193.1 69.160.2:53682 <-> 193.169.160.2:53684 <-> 193.169.160.50:9164 (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:39 earth media-relay[10532]: debug: Got traffic information for stream: (audio) 193.169.160.50:8554 (RTP: Unknown, RTCP: Unknown) <-> 193.169.160. 2:53678 <-> 193.169.160.2:53680 <-> 193.169.160.50:9064 (RTP: 193.169.160.50:9064, RTCP: Unknown) Aug 6 09:15:40 earth media-relay[10532]: debug: Got traffic information for stream: (audio) 193.169.160.50:8554 (RTP: 193.169.160.50:8554, RTCP: Unknown) <-> 193.169.160.2:53678 <-> 193.169.160.2:53680 <-> 193.169.160.50:9064 (RTP: 193.169.160.50:9064, RTCP: Unknown) Aug 6 09:15:41 earth media-relay[10532]: debug: Got traffic information for stream: (video) 193.169.160.50:8458 (RTP: 193.169.160.50:8458, RTCP: Unknown) <-> 193.169.160.2:53682 <-> 193.169.160.2:53684 <-> 193.169.160.50:9164 (RTP: Unknown, RTCP: Unknown) Aug 6 09:15:46 earth media-relay[10532]: debug: Got traffic information for stream: (video) 193.169.160.50:8458 (RTP: 193.169.160.50:8458, RTCP: Unknown) <-> 193.169.160.2:53682 <-> 193.169.160.2:53684 <-> 193.169.160.50:9164 (RTP: 193.169.160.50:9164, RTCP: Unknown) Aug 6 09:16:40 earth media-relay[10532]: (Port 53678 Closed) Aug 6 09:16:40 earth media-relay[10532]: (Port 53679 Closed) Aug 6 09:16:40 earth media-relay[10532]: (Port 53680 Closed) Aug 6 09:16:40 earth media-relay[10532]: (Port 53681 Closed) Aug 6 09:16:56 earth media-relay[10532]: debug: removing session 1a632798455277b8: 93173...@tipmeet.com (ddd06077a660b1dc) --> di...@tipmeet.com Aug 6 09:16:56 earth media-relay[10532]:
[OpenSIPS-Users] Fedora package for 1.6.1
Hi all.
I compiled opensips-1.6.1 rpm package for fedora 12 and rawhide:
http://kojipkgs.fedoraproject.org/packages/opensips/1.6.1/1.fc12/
http://kojipkgs.fedoraproject.org/packages/opensips/1.6.1/1.fc13/
Also package was pushed to the updates-testing:
https://admin.fedoraproject.org/updates/opensips-1.6.1-1.fc12
Welcome to testing!
To bogdan: spec file changes from opensips upstream to fedora upstream
attached to mail. Apply changes, please.
--
WBR, John Khvatov
diff --git a/opensips.spec?revision=6456 b/opensips.spec
index eb36c18..7272cab 100644
--- a/opensips.spec?revision=6456
+++ b/opensips.spec
@@ -10,13 +10,11 @@
Summary: Open Source SIP Server
Name: opensips
Version: 1.6.1
-Release: 4%{?dist}
+Release: 1%{?dist}
License: GPLv2+
Group:System Environment/Daemons
Source0:
http://opensips.org/pub/%{name}/%{version}/src/%{name}-%{version}-tls_src.tar.gz
Source1: %{name}.sysconfig
-Patch1: opensips--init.patch
-Patch2: opensips--openssl10.patch
URL: http://opensips.org
BuildRequires: expat-devel
@@ -502,8 +500,6 @@ clients.
%prep
%setup -q -n %{name}-%{version}-tls
-%patch1 -p1
-%patch2 -p1
%build
LOCALBASE=/usr CFLAGS="%{optflags}" %{__make} all %{?_smp_mflags} TLS=1 \
@@ -990,6 +986,10 @@ fi
%doc docdir/README.xmpp
%changelog
+* Thu Dec 22 2009 John Khvatov - 1.6.1-1:
+- Updated to 1.6.1
+- Dropped upstreamed patches
+
* Wed Nov 04 2009 John Khvatov - 1.6.0-4:
- Fixed typo: pia_mi to pua_mi in presence_xcapdiff dependencies
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] patch for compatibility with openssl-1.0.0
Hello all!
Fedora 12 will be with new OpenSSL. Could you apply attached patch for
fine build with new OpenSSL?
Current version of the code fails to build:
tls/tls_init.c: In function 'check_for_krb':
tls/tls_init.c:470: error: 'struct stack_st_SSL_CIPHER' has no member
named 'num'
tls/tls_init.c:471: error: 'struct stack_st_SSL_CIPHER' has no member
named 'data'
make: *** [tls/tls_init.o] Error 1
--
WBR, John Khvatov
diff -up opensips-1.6.0/tls/tls_init.c.openssl opensips-1.6.0/tls/tls_init.c
--- opensips-1.6.0/tls/tls_init.c.openssl 2008-11-24 21:52:33.0
+0100
+++ opensips-1.6.0/tls/tls_init.c 2009-08-25 20:18:45.0 +0200
@@ -468,8 +468,8 @@ static int check_for_krb(void)
if (xx==NULL)
return -1;
- for( j=0 ; jcipher_list) ; j++) {
- SSL_CIPHER *yy = (SSL_CIPHER*)M_sk_value(xx->cipher_list,j);
+ for( j=0 ; jcipher_list) ; j++) {
+ SSL_CIPHER *yy = sk_SSL_CIPHER_value(xx->cipher_list,j);
if ( yy->id>=SSL3_CK_KRB5_DES_64_CBC_SHA &&
yy->id<=SSL3_CK_KRB5_RC4_40_MD5 ) {
LM_INFO("KRB5 cipher %s found\n", yy->name);
___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] fedora rpm package
Hello all. I'm opensips maintainer on fedora. This patch for init script to fix malformed comment block: http://dev.sgu.ru/fedora/opensips--init.patch Current spec file with new opensips modules: http://dev.sgu.ru/fedora/opensips.spec Consider to apply these changes to the upstream, please. -- WBR, John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to run opensips in the foreground
Hi Bogdan, On Mon, Apr 06, 2009 at 10:02:49PM +0300, Bogdan-Andrei Iancu wrote: > Hi John, > > i'm using monit (see http://mmonit.com/monit/ ) for a long time - quite > nice; > > Also you may consider nagios http://www.nagios.org/ > > Regards, > Bogdan Thanks. -- John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to run opensips in the foreground
On Fri, Apr 03, 2009 at 11:27:31PM +0400, John Khvatov wrote: > On Fri, Apr 03, 2009 at 08:26:09PM +0200, Iñaki Baz Castillo wrote: > > > > If you set "log_stderror=no" then OpenSIPS output will be logged in the > > terminal. But you can close safely this terminal session and OpenSIPS will > > remain working. > > > > BTW, which requeriments has upstart? > > Upstart until release 5.0 doesn't support processes that detach from terminal > (that fork (or fork twice)). Upstart works with opensips if I set "fork=no" > in opensips config. But this way is not suitable for prodaction, right? > There are other ways to run opensips foreground? Upstart 0.5, sorry. -- John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
Re: [OpenSIPS-Users] How to run opensips in the foreground
On Fri, Apr 03, 2009 at 08:26:09PM +0200, Iñaki Baz Castillo wrote: > > If you set "log_stderror=no" then OpenSIPS output will be logged in the > terminal. But you can close safely this terminal session and OpenSIPS will > remain working. > > BTW, which requeriments has upstart? Upstart until release 5.0 doesn't support processes that detach from terminal (that fork (or fork twice)). Upstart works with opensips if I set "fork=no" in opensips config. But this way is not suitable for prodaction, right? There are other ways to run opensips foreground? -- John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
[OpenSIPS-Users] How to run opensips in the foreground
Hello! Can I run opensips in the foreground in "daemon mode" - multiple processes for each network interface? I need it to manage the opensips service by upstart [1]. [1] http://upstart.ubuntu.com/ Thanks. -- John Khvatov ___ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
