Stefan Sayer writes:

> looks like this happens when destroying the plugins. which ones do you 
> have loaded? Does that also happen when no call is running? Can you 
> try to insert some more DEBUG info in delete_plugin_factory in 
> AmPlugIn.cpp:115 ?

i turned the debug in delete_plugin_factory to info and got:

May 16 07:55:25 sip sems[12979]: [#b754d6d0] [main, sems.cpp:509] INFO: 
Disposing plug-ins
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: announcement
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: conference
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: early_announce
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: echo
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: sbc
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: session_timer
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: stats
May 16 07:55:26 sip sems[12979]: [#b754d6d0] [delete_plugin_factory, 
AmPlugIn.cpp:118] INFO: deleting plug-in factory: user_timer
May 16 07:55:26 sip kernel: [30165.929795] sems[12981]: segfault at b723b284 ip 
b723b284 sp b6a9baf0 error 4 in ivr.so[b7240000+1a000]

when sems is not crashing i get instead of the last entry:

May 16 07:56:27 sip sems[26511]: [#b75f36d0] [main, sems.cpp:520] INFO: Exiting 
(success)

so looks like the crash happens while or after deleting the last, i.e.,
user_timer, plugin.

there were no calls active when crash happened.  this is on my laptop
and crash happened when i restarted sems after resuming from
hibernation.  i don't know yet, if it happens also without hibernation.
anyway, this kind of crash has never happened to me in 1.2 version of
sems.

-- juha
_______________________________________________
Sems mailing list
[email protected]
http://lists.iptel.org/mailman/listinfo/sems

Reply via email to