Asterisk not response SIP OPTIONS

Good morning, I have this problem that is becoming more recurrent, an asterisk server stops responding messages “sip options” from other servers and users, so the other sip servers see asterisk dropped at the SIP level, I am using Asterisk certified /16.8-cert9 with the PJSIP channel, however the asterisk service continues to run.

Someone else has had this problem and they have seen some way to fix it, or if it is a configuration that you should modify.

Thanks.

By enabling DEBUG, I get these messages.
[Oct 5 12:40:02] DEBUG[7979]: res_pjsip/pjsip_distributor.c:503 distributor: Searching for serializer associated with dialog dlg0x7f379823d348 for Response msg 481/ACK/cseq=15893 (rdata0x7f37852a0f38)
[Oct 5 12:40:02] DEBUG[7979]: res_pjsip/pjsip_distributor.c:520 distributor: No dialog serializer for Response msg 481/ACK/cseq=15893 (rdata0x7f37852a0f38). Using request transaction as basis.
[Oct 5 12:40:02] DEBUG[7979]: res_pjsip/pjsip_distributor.c:128 find_request_serializer: Found transaction tsx0x7f3798e68298 for Response msg 481/ACK/cseq=15893 (rdata0x7f37852a0f38).
[Oct 5 12:40:02] DEBUG[7979]: res_pjsip/pjsip_distributor.c:138 find_request_serializer: Found serializer pjsip/distributor-000026e9 on transaction tsx0x7f3798e68298

When the asterisk interconnected to this via sip trunk send options, they do not get responses, although capturing traffic with sngrep I see that if the sip options arrive, IP ASTERISK WITH PROBLEM IS 192.168.0.5
options

the only solution I can find at the moment is to restart asterisk so that it works again.

I see this happening 2 to 3 times a week.

I would appreciate if anyone has a similar case that they have been able to solve or support to solve this.

Which version of Asterisk?

I’d suggest diagnosing this on the assumption that it is a deadlock: Getting a Backtrace - Asterisk Project - Asterisk Project Wiki

My Versión is: Asterisk certified /16.8-cert9

I understand that a backtrace works when the system crashes or am I wrong? if so, as I said lines above, the system never crashes, it is the SIP OPTIONS that arrive that are never answered, causing the established sip trunks to fall from the remote side. however asterisk continues to work apparently, so the restart must be done manually.

The only reason for having a certified version is that you have a support contract from Sangoma, in which case you should use that. Otherwise you should be on the latest community release of Asterisk 16.

Core dumps can be created on running processes. Also, you said you had to restart Asterisk, in which case you could consider bringing it down with kill -3 to trigger crash handling.

Thanks for your comments.

During these days this has not happened again, however I have updated to the latest version of asterisk from branch 16, Asterisk 16.21.0 and took the opportunity to enable DONT OPTIMIZE when compiling asterisk to obtain the backtrace in case it happens again.

I’ll keep you posted when this issue happens again.

Thank you.

Today this problem happened again, on the side of my asterisk PBX I always see the active trunks, as in the image, however, on the side of the other PBXs they send me options and my PBX never responds to them, whether they are PBX at level LAN or over WAN.
image

And this is how my PBX looks like it does not respond to the sip options from the other PBXs

Today 15 Oct 2021 I was able to run the command:

/var/lib/asterisk/scripts/ast_coredumper --running

While the asterisk server was running with this problem and I got the following files:

They can help me by indicating which of all of them I should upload them so that they can support me by analyzing them and being able to find the error and any possible solution for this.

Thanks.

hi, does anyone have any advice to fix this?