I am using FreePBX 2.10.0-99 with Asterisk 18.104.22.168-34_centos6.
I have had problems that at times the phones would not work at all. They would wait for some time before initiating a call. I put sip into debug mode and discovered a number of entries similar to:
[2013-03-11 15:59:38] VERBOSE chan_sip.c: --- (8 headers 0 lines) ---
[2013-03-11 15:59:38] VERBOSE chan_sip.c: Responding to challenge, registration to domain/host name myprovider.org
[2013-03-11 15:59:38] VERBOSE dnsmgr.c: > doing dnsmgr_lookup for 'myprovider.org'
[2013-03-11 15:59:58] VERBOSE chan_sip.c: REGISTER 11 headers, 0 lines
[2013-03-11 15:59:58] VERBOSE chan_sip.c: Reliably Transmitting (NAT) to XXX.XXX.XXX.XXX:5060:
REGISTER sip:myprovider.org SIP/2.0
What I noticed is that between 15:59:38 and 15:59:58 nothing happens on the system while the the dnsmgr_lookup is happening and this is the period when phones do not work at all. It seems that the lookup times out after 20 seconds and the system goes back to working successfully.
I have circumvented this problem by adding “srvlookup=no” to sip.conf. I am still slightly concerned that I may need srvlookup some time and so would like to resolve the underlying problem. I am planning to do some network tracing to see what is happening, but before I embark on that I would like to check that I am working in the correct direction.
One thing that I have noticed is that ping to my service provider never completes:
PING myprovider.org (XXX.XXX.XXX.XXX) 56(84) bytes of data.
--- myprovider.org ping statistics ---
19 packets transmitted, 0 received, 100% packet loss, time 18652ms
and I was wondering whether dnsmgr_lookup was trying a ping and then timing out. Is that likely to be related?