[asterisk-bugs] [JIRA] (ASTERISK-22208) Extremely high CPU load and segfaults (in ast_hashtab_start_traversal) on a reload

[David Brillert (JIRA)]

     [ https://issues.asterisk.org/jira/browse/ASTERISK-22208?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

David Brillert updated ASTERISK-22208:
--------------------------------------

    Attachment: gdb thread apply all bt 8192013.txt
                core show locks 8192013.txt

I ran core show locks and gdb thread commands when I thought Asterisk was locking during a reload.  In other words when I see Asterisk reload hanging just after  

[2013-08-19 09:34:24]   == Parsing '/etc/asterisk/users.conf': Found
[2013-08-19 09:34:24]     -- Registered extension context 'parkedcalls'; registrar: features
[2013-08-19 09:34:24]     -- merging incls/swits/igpats from old(parkedcalls) to new(parkedcalls) context, registrar = pbx_config
[2013-08-19 09:34:24]     -- Added extension '7000' priority 1 to parkedcalls
[2013-08-19 09:34:24]     -- merging incls/swits/igpats from old(parkedcalls_default) to new(parkedcalls_default) context, registrar = pbx_config
[2013-08-19 09:34:24] WARNING[4007]: pbx.c:9619 add_priority: Unable to register extension '7000' priority 1 in 'parkedcalls_default', already in use
[2013-08-19 09:34:24]     -- Dropping old dialplan item parkedcalls_default/7000/1 [Park()] (registrar=features) due to conflict with new dialplan
[2013-08-19 09:34:24]     -- merging incls/swits/igpats from old(parkedcalls_tenant) to new(parkedcalls_tenant) context, registrar = pbx_config
[2013-08-19 09:34:24] WARNING[4007]: pbx.c:9619 add_priority: Unable to register extension '700' priority 1 in 'parkedcalls_tenant', already in use
[2013-08-19 09:34:24]     -- Dropping old dialplan item parkedcalls_tenant/700/1 [Park()] (registrar=features) due to conflict with new dialplan
demo*CLI> core show locks


I had also compiled with BETTER_BACKTRACES DONT_OPTIMIZE DEBUG_THREADS and I'm waiting for another segfault to provide the requested core dump output.

Core and gdb output is
core show locks 8192013.txt
gdb thread apply all bt 8192013.txt
                
> Extremely high CPU load and segfaults (in ast_hashtab_start_traversal) on a reload
> ----------------------------------------------------------------------------------
>
>                 Key: ASTERISK-22208
>                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-22208
>             Project: Asterisk
>          Issue Type: Bug
>      Security Level: None
>          Components: Channels/General, Core/Configuration
>    Affects Versions: 11.5.0
>         Environment: CentOS 5.9 32 bit
>            Reporter: David Brillert
>            Severity: Critical
>         Attachments: 7_26_2013 gdb thread apply all bt full.txt, core 12843 segfault bt full.txt, core_31T09_gdb bt full.txt, core show locks 8192013.txt, core_show_locks.txt, extensions.rar, features.conf, gdb thread apply all bt 8192013.txt, lab_7_31_2013 gdb thread apply all bt full.txt, messages.rar, valgrind_july_31.rar, valgrind.rar
>
>
> CPU load in htop will register in the 60's and or 80's after a reload until I restart Asterisk.
> Ticket is open at the request of Rusty Newton in relationship to ASTERISK-22093

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.asterisk.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira