[sr-dev] a new master crash in tm module

Daniel-Constantin Mierla miconda at gmail.com
Wed Sep 23 10:19:36 CEST 2015


I am investigating. Can you get the following:

frame 0
info locals
p *qm
p *frag
frame 1
info locals
p *f
p *n

Thanks,
Daniel

On 23/09/15 10:15, Juha Heinanen wrote:
> Daniel-Constantin Mierla writes:
>
>> Have you start it with -x qm or without?
>>
>> >From the core of the next bt, can you get:
>>
>> frame 0
>> p *t
>> p *p_msg
>> info locals
> when i start k with -x qm, i have not managed to get a crash.  however,
> if i start k without -x qm, it crashes reliably, for example:
>
> (gdb) where
> #0  0x0000000000641acc in fm_extract_free (qm=0x7f28c8dd4000, 
>     frag=0x7f28c958e4a0) at mem/f_malloc.c:181
> #1  0x0000000000643689 in fm_join_frag (qm=0x7f28c8dd4000, f=0x7f28c958e468)
>     at mem/f_malloc.c:556
> #2  0x00000000006445ac in fm_free (qmp=0x7f28c8dd4000, p=0x7f28c958e4a0, 
>     file=0x7f28ce687f1d "tm: h_table.c", 
>     func=0x7f28ce6881bb <__FUNCTION__.9593> "free_cell", line=162)
>     at mem/f_malloc.c:624
> #3  0x00007f28ce5c1c12 in free_cell (dead_cell=0x7f28c95eb6c0) at h_table.c:162
> #4  0x00007f28ce64449e in wait_handler (ti=1463515017, wait_tl=0x7f28c95eb740, 
>     data=0x7f28c95eb6c0) at timer.c:648
> #5  0x00000000004aaf88 in timer_list_expire (t=1463515017, h=0x7f28c8e21360, 
>     slow_l=0x7f28c8e21e38, slow_mark=153) at timer.c:873
> #6  0x00000000004ab3e5 in timer_handler () at timer.c:938
> #7  0x00000000004ab853 in timer_main () at timer.c:977
> #8  0x000000000052f690 in main_loop () at main.c:1650
> #9  0x00000000005354e8 in main (argc=17, argv=0x7ffcfb5ef7a8) at main.c:2566
>
> -- juha

-- 
Daniel-Constantin Mierla
http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
Book: SIP Routing With Kamailio - http://www.asipto.com
Kamailio Advanced Training, Sep 28-30, 2015, in Berlin - http://asipto.com/u/kat




More information about the sr-dev mailing list