[OpenSIPS-Devel] [opensips] SIGSEGV in pkg_malloc/fm_malloc (#721)

Stéphane Alnet notifications at github.com
Mon Dec 14 15:58:11 CET 2015


@razvancrainea  yes, we had the issue maybe every hour or so (on one server) on the day it first appeared (we also had it on other servers with a lesser rate). I hadn't done much work with the internals of OpenSIPS so didn't know about the memory debugging allocator, and since it was a SIGSEGV I first tried to figure out where the crash was happening; it's only later that I figured out that the memory allocator was involved and that the actual issue might be incorrect use of the allocator.

We had some low-rate unwanted traffic on those servers while this was happening (INVITEs, with no ACKs) and the problems kind-of disappeared after I blocked the IPs that were causing trouble; however I can't say for sure this was the source of the problem; either way this might be difficult to re-create.

Following up on @liviuchircu 's note, the only `CRIT`-level notices I see on the new versions (with #722 applied, running over the week-end) are `CRITICAL:tm:set_timer: set_timer for 1 list called on a "detached" timer -- ignoring`.
→ Do I need to set `memlog=1` for this to work?

I'll see if I can build a test image with the memory allocator and put it alongside the production images.
→ should I use the QM allocator (`DBG_QM_MALLOC`), or is it enough to use `DBG_F_MALLOC` ?

I guess I could also `abort()` at the point in the allocator where the problem is detected.

---
Reply to this email directly or view it on GitHub:
https://github.com/OpenSIPS/opensips/issues/721#issuecomment-164458308
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/devel/attachments/20151214/f7fed58f/attachment.htm>


More information about the Devel mailing list