Re: [SR-Users] Possible bug in TM module

Hi, For now, I got it only once, I will keep an eye on it. Thanks, Mickael From: sr-users [mailto:sr-users-bounces@lists.sip-router.org] On Behalf Of Daniel-Constantin Mierla Sent: Wednesday, April 22, 2015 12:37 PM To: Kamailio (SER) - Users Mailing List Subject: Re: [SR-Users] Possible bug in TM module Hello, On 22/04/15 11:26, Mickael Marrache wrote: Hi, I just got the following issue when stopping kamailio. kamailio[7279]: NOTICE: <core> [main.c:682]: handle_sigs(): Thank you for flying kamailio!!! kamailio[7279]: ERROR: ctl [ctl.c:369]: mod_destroy(): ERROR: ctl: could not delete unix socket /tmp/kamailio_ctl: Operation not permitted (1) kamailio[7279]: : <core> [mem/f_malloc.c:586]: fm_free(): BUG: fm_free: bad pointer 0x4f2dec6200000000 (out of memory block!), called from tm: h_table.c: free_cell(138) - aborting I always get the first ERROR when I stop kamailio, I need to look at the permissions. However, the next one is a bug. the first one should be fixed with the new release. With old versions, ctl module doesn't inherits the user/group set by the core, you would have to set it in the ctl module via params. The second can happen in very rare cases -- if there was a transaction in the process of freeing, then signal interrupted it. Parts of the transaction structure that were freed could have been allocated already and value overwritten. But the transaction itself was not removed from the list, so the main process will attempt a cleanup at shutdown, ending in some invalid pointer data. If you get it every time, then is likely a bug. If you get it very rare, then it can be the above case. Perhaps we can add a flag saying it is shutdown where to be more flexible in handling these situations. Cheers, Daniel -- Daniel-Constantin Mierla http://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda Kamailio World Conference, May 27-29, 2015 Berlin, Germany - http://www.kamailioworld.com