Hi Bogdan,
It doesnt seem to have fixed the memory problem. Actually this memory error
also occured in the previous version I was running of "openser
1.1.0-dev17-notls". Also note the next_gw lcr error.
Jun 26 17:50:11 sipserver01 /sbin/openser[15337]: ERROR: new_t: out of mem:
Jun 26 17:50
Hi Sumeet,
have you tried to use the latest CVS head? I made a small fix couple of
hours ago...please update and see if the problem persists.
regards,
bogdan
Sumeet Gupta wrote:
I believe I am experiencing the same problem. Increasing the shared
memory size does not help either. I increased
I believe I am experiencing the same problem. Increasing the shared memory
size does not help either. I increased it to 128 megs with high load and
still only moments after startup i recieve the following errors.
-Sumeet
Jun 26 11:44:58 sipserver01 /sbin/openser[21682]: ERROR: sip_msg_cloner:
Andreas,
the log shows that some transactions cannot be deleted because some
referencing problem, I suspect it is somehow related to the ACK fix. I
did a small fixup on CVS and added some more debug in REF/UNREF to try
to figure out what is the problem.
So, first step is to update from cvs a
User: bogdan_iancu
Date: 2006/06/26 10:59:03 PDT
OpenSER CVS - Commit Details
Modified files:
modules/tm t_funcs.c t_funcs.h t_lookup.c timer.c
Commit Log:
- small fixup in ACK processing.
- more debug added for troubleshooting some memory leak.
Revision Changes
Hi Andreas,
could you please send me privately the shm mem log during run time
(SIGUSR1) and at shutdown? I will take a look and see if I can get
something out of it
what it will be very, very , very helpful is the complete log (from
startup to shutdown) of openser with mem debug support
thanks,
razvan radu
Bogdan-Andrei Iancu wrote:
Salut Razvan,
yes that's right, the comments in the sip_msg structure are misleading
and the FL_REQ_UPSTREAM flag no not set by ACC module in the proper
flags set.
I fixed it on the CVS.
thanks and regards,
Bogdan
[EMAIL PROTECTED] wrote:
User: agranig
Date: 2006/06/26 07:32:31 PDT
OpenSER CVS - Commit Details
Modified files:
modules/path path.c
Commit Log:
Fixed mem-leak in parsing route-params.
Revision ChangesPath
1.3 +3 -1 sip-server/modules/path/path.c
[
http://openser.cvs.sour
Bugs item #1511976, was opened at 2006-06-25 03:49
Message generated for change (Comment added) made by bogdan_iancu
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1511976&group_id=139143
Please note that this message will contain a full copy of the comm
User: bogdan_iancu
Date: 2006/06/26 07:28:08 PDT
OpenSER CVS - Commit Details
Modified files:(Branch: rel_1_0_0)
.Makefile
Commit Log:
addition to the original fix:
- "make all" compile also the utils (like gen_ha1) to avoid permission
problems
dur
I wrote:
However, I'll wait half an hour and check the memlog again...
Same thing after half an hour. The shared memory allocated by shm_resize
still wasn't free'd. Any ideas?
On the other hand, I found just another pkg-mem-leak in the path module.
I'll commit a fix asap.
Andy
_
Bugs item #1512700, was opened at 2006-06-26 16:11
Message generated for change (Comment added) made by bogdan_iancu
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1512700&group_id=139143
Please note that this message will contain a full copy of the comm
Hi,
Bogdan-Andrei Iancu wrote:
it's not necessary a mem leak..please see the
http://openser.org/dokuwiki/doku.php?id=memory
go through and let me know the result. It might be just a memory
starvation due high load.
What makes me suspicious is the fact that the number of calls I can
establi
Bugs item #1512700, was opened at 2006-06-26 13:11
Message generated for change (Tracker Item Submitted) made by Item Submitter
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1512700&group_id=139143
Please note that this message will contain a full copy
Hi Andreas,
it's not necessary a mem leak..please see the
http://openser.org/dokuwiki/doku.php?id=memory
go through and let me know the result. It might be just a memory
starvation due high load.
regards,
bogdan
Andreas Granig wrote:
Hi,
Seems like there's another mem-leak in OpenSER (
Hi,
Seems like there's another mem-leak in OpenSER (CVS HEAD). If I simulate
thousands of calls, the proxy runs out of shared memory, with
memlog-lines like the following at shutdown:
57583. N address=0xb77a87c0 frag=0xb77a87a8 size=888 used=1
alloc'd from mem/shm_mem.c: sh_realloc
Hi,
the change is on the CVS.
thanks for report!
regards,
Bogdan
Bogdan-Andrei Iancu wrote:
Hi Mike,
I have to agree with you - I guess the best way to go will be to have
anew field as unique id (autoincrement) to be used as primary key -
the way the AVPs are defined makes very complicate
G.Jacobsen wrote:
I note that user_preferences has uuid as start of the primary key - up
to version 1.0.1 the username was the key.
Surely I dont like to cause any offence with the developers - but
wouldnt it be preferable to have consistent key indexes across all
modules for similar functio
User: bogdan_iancu
Date: 2006/06/26 01:48:56 PDT
OpenSER CVS - Commit Details
Modified files:
scripts mysqldb.sh postgresqldb.sh
Commit Log:
- fix issue with primary key in usr_preferences (AVP) table.
Credits go to Mike Williams <[EMAIL PROTECTED]> and
G.Jacobs
User: bogdan_iancu
Date: 2006/06/26 01:42:26 PDT
OpenSER CVS - Commit Details
Modified files:
.resolve.c
Commit Log:
- compiling warning fixed
Revision ChangesPath
1.11 +4 -3 sip-server/resolve.c
[
http://openser.cvs.sourceforge.net/opens
Salut Razvan,
yes that's right, the comments in the sip_msg structure are misleading
and the FL_REQ_UPSTREAM flag no not set by ACC module in the proper
flags set.
I fixed it on the CVS.
thanks and regards,
Bogdan
[EMAIL PROTECTED] wrote:
hello,
please ignore my message if this is the in
User: bogdan_iancu
Date: 2006/06/26 01:37:13 PDT
OpenSER CVS - Commit Details
Modified files:
modules/acc acc.c acc_mod.c
Commit Log:
- fixed bug - the internal and script sets of flags whre swapped when
using the FL_REQ_UPSTREAM flag
Credits go to razvan radu <[EMAI
User: bogdan_iancu
Date: 2006/06/26 01:35:34 PDT
OpenSER CVS - Commit Details
Modified files:
parser msg_parser.h
Commit Log:
- fixed misleading comments regarding the internal and script sets of flags.
Reported by razvan radu <[EMAIL PROTECTED]>
Revision Chan
Hi Andreas,
that's great! thanks for the fix.
regards,
bogdan
Andreas Granig wrote:
I wrote:
What I'm looking for is the place where the "struct sip_msg *msg" is
free'd when processing the request has finished. I assume that
path_vec.s is left allocated there.
Ok, found it, it's free_si
Hi Ray,
it looks like the "trusted" hash table is corrupted. If I understand
correctly, you say if you start the proxy and send an INVITE, the proxy
will immediately crash, right? do you see any errors during startup or
before the crash?
regards,
bogdan
Raymond Chen wrote:
Hi Bogdan,
yes
Summet,
yes, indeed - with the ACK fix I introduced a small bug. It is fixed now
on the CVS - please try to update and test again.
thanks and regards,
bogdan
Sumeet Gupta wrote:
Bogdan,
Thanks that fixed that particular error, but I was getting other
errors regarding calculation of ACK br
User: bogdan_iancu
Date: 2006/06/26 01:20:01 PDT
OpenSER CVS - Commit Details
Modified files:
.resolve.c
Commit Log:
- fixed small potential memory leak in SIP resolver.
Reported by Douglas Garstang <[EMAIL PROTECTED]>
Revision ChangesPath
1.10
User: bogdan_iancu
Date: 2006/06/26 01:18:54 PDT
OpenSER CVS - Commit Details
Modified files:
modules/tm t_funcs.c
Commit Log:
- fixed bug introduced with the latest ACK fixup.
reported by Andreas Granig <[EMAIL PROTECTED]>
Revision ChangesPath
1.15 +17
28 matches
Mail list logo
