[Users] troubles with usrloc module: db_mode=3 (DB only)

Christian Schlatter cs at unc.edu
Thu Jun 8 16:51:16 CEST 2006


Bogdan-Andrei Iancu wrote:
> Hi Christian,
> 
> getting a backtrace will be very, very helpful. be sure the you are 
> using exactly the same binaries as the ones generating the core file.
> 
> Also, could you describe the scenario leading to the crash? also set 
> debug=9 and the log will be very useful.

Ok, I got some more insights now (I took the wrong binary for the 
backtrace). The whole issue seems to be related to the nat_pinging 
function of the mediaproxy module in combination with db_mode=3. But 
first the backtrace:

#0  0x00b2b392 in get_all_db_ucontacts (buf=0x8135e20, len=252, flags=1) 
at dlist.c:150
#1  0x002cfcf0 in pingClients (ticks=30, param=0x0) at functions.h:44
#2  0x08088afb in timer_ticker () at timer.c:141
#3  0x08063e94 in main_loop () at main.c:988
#4  0x0806504c in main (argc=7, argv=0xbfff8de4) at main.c:1473

And these are my nat_pinging settings:

# -- nathelper params --
modparam("nathelper", "rtpproxy_disable", 1)
modparam("nathelper", "natping_interval", 0)

# -- mediaproxy params --
modparam("mediaproxy","natping_interval", 30)


If I disable the nat pinging by setting the interval 0 the server just 
runs fine with db_mode=3. I haven't tested yet if the nathelper 
nat_pinging causes the same issues as the mediaproxy one.

This is what the debug=9 log tells (I filtered the log for messages from 
the process the crashes):

[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=tm
[21065]: DEBUG: callid: '7fbb7eae-21065 at 152.2.19.138'
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=usrloc
[21065]: db_init: Connection 
'mysql://openser:openserrw@localhost/openser' not found in pool
[21065]: new_connection: Opening MySQL connection: 
mysql://openser:openserrw@localhost/openser
[21065]: new_connection: Connection type is Localhost via UNIX socket
[21065]: new_connection: Protocol version is 10
[21065]: new_connection: Server version is 3.23.58-classic-log
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=alias_db
[21065]: db_init: Connection 
'mysql://openser:openserrw@localhost/openser' found in pool
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=auth_db
[21065]: db_init: Connection 
'mysql://openserro:openserro@localhost/openser' not found in pool
[21065]: new_connection: Opening MySQL connection: 
mysql://openserro:openserro@localhost/openser
[21065]: new_connection: Connection type is Localhost via UNIX socket
[21065]: new_connection: Protocol version is 10
[21065]: new_connection: Server version is 3.23.58-classic-log
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=uri_db
[21065]: db_init: Connection 
'mysql://openser:openserrw@localhost/openser' found in pool
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=domain
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=nathelper
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=avpops
[21065]: db_init: Connection 
'mysql://openser:openserrw@localhost/openser' found in pool
[21065]: DEBUG:init_mod_child: PROC_TIMER , rank=-1, module=xlog
[21065]: XLOG: init_child [-1]  pid [21065]
[21070]: DBG: handle_ser_child: dead child 6, pid 21065 (shutting down?)
[21054]: child process 21065 exited by a signal 11
[21054]: core was generated


Thanks a lot for looking at that,
Christian


> regards,
> bogdan
> 
> Christian Schlatter wrote:
> 
>> Hi all,
>>
>> Running openser (CVS HEAD) with usrloc:db_mode=3 causes it to crash 
>> repeatedly after 15 secs or so. The core dump does not say much except
>>
>> (gdb) bt
>> #0  0x00b2b392 in ?? ()
>> #1  0x08131468 in mem_pool ()
>> #2  0xbfff8b54 in ?? ()
>> #3  0xbfff8b54 in ?? ()
>> #4  0x00000008 in ?? ()
>> #5  0x00b3adbf in ?? ()
>> #6  0x00000007 in ?? ()
>> #7  0x00b3a88c in ?? ()
>> #8  0x00000006 in ?? ()
>> #9  0x00b3adcd in ?? ()
>> #10 0x00000005 in ?? ()
>> #11 0x00b3adae in ?? ()
>> #12 0xb576a560 in ?? ()
>> #13 0x00000007 in ?? ()
>> #14 0x00b39c59 in ?? ()
>> #15 0x00000005 in ?? ()
>> #16 0x00b3adae in ?? ()
>> #17 0x00000001 in ?? ()
>> #18 0x00000001 in ?? ()
>> #19 0x00000000 in ?? ()
>>
>>
>> The openser log looks like:
>>
>> ERROR: receive_fd: EOF on 16
>> child process 18101 exited by a signal 11
>> core was generated
>>
>>
>> I also don't see any mysql errors. The last mysql query issued by 
>> openser before dying is always
>>
>> Query  delete from location where expires<'2006-06-05 16:03:17'
>> (only the timestamp changes)
>>
>> Deleting all location table entries before starting up openser does 
>> not help either. Is there something special I have to consider when 
>> running openser in DB-only mode?
>>
>> Thanks for any help,
>> Christian
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> Users mailing list
>> Users at openser.org
>> http://openser.org/cgi-bin/mailman/listinfo/users
>>  
>>
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 1951 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://lists.sip-router.org/pipermail/sr-users/attachments/20060608/1332cb1e/attachment.bin>


More information about the sr-users mailing list