[asterisk-dev] Suspected deadlocks in Asterisk 1.8 under heavy load

[Alistair Cunningham]

Just to let everyone know, we strongly suspect that Asterisk 1.8.4.4 and 
1.8.5.0 suffer from occasional deadlocks under heavy load, possibly 
related to the local channel. The symptoms seem to vary, but include:

1. Asterisk stops responding to SIP packets, but the Asterisk console 
and manager interface remain responsive. A "sipsak -s sip:127.0.0.1:5060 
-d" times out.

2. The console stops responding to "core show channels":

asterisk01*CLI> core show channels
Channel Location State Application(Data)
asterisk01*CLI> core show channels
asterisk01*CLI> core show channels
asterisk01*CLI> quit

Connecting with a new console then does the same.

3. Lots of the following:

[Aug 15 14:28:08] WARNING[2158]: chan_local.c:204 local_setoption: Could 
not update other side of Local/6 at enswitch-call-exten-3b6e;2, other side 
went away.

4. Lots of the following:

[Aug 15 14:14:32] WARNING[4038]: channel.c:1504 __ast_queue_frame: 
Unable to write to alert pipe on Local/1234567890 at enswitch-local-2227;1 
(qlen = 53): Resource temporarily unavailable!

This is on a system with call pickup disabled, so seems to be unrelated 
to the call pickup problems.

We will continue investigating, but our primary response for any 
affected customers is going to be to downgrade them to 1.6.1 which we're 
confident is stable. We're probably also going to suspend upgrades to 1.8.

-- 
Alistair Cunningham
+1 888 468 3111
+44 20 799 39 799
http://integrics.com/