There is a delay of 5+ seconds when I place an outbound call from an IP phone (both soft phones and hard phones) over either a sip or iax trunk.
When I place a call this appears in the logs immediately:
== Using SIP RTP CoS mark 5
– Called SIP/voipmstrunksip/15182097178
Then 5+ seconds later:
– SIP/voipmstrunksip-00000020 is making progress passing it to SIP/1001-0000001f
> 0x7f8c6013c630 – Probation passed - setting RTP source address to 107.6.67.238:17532
> 0x7f8c5807af00 – Probation passed - setting RTP source address to 74.70.48.112:7078
> 0x7f8c5807af00 – Probation passed - setting RTP source address to 74.70.48.112:7078
Funny thing is when the PBX itself routes a call over the trunk instead of an IP phone initiating the call, it goes through immediately. For example I have one inbound route for a DID that goes into a ring group. The ring group contains my cell phone. When I call the DID from an outside line, the PBX initiates an outgoing call over the sip/iax trunk to my cell phone and it happens immediately, no 5+ second delay like when using an IP phone.
It is an inefficient “dialstring” on your phone, if there is no complete match, then the default timeout is generally 5 seconds, you can normally prove that by adding a # (dial string terminator) after the dialed numbers.
I have been using linphone to do my testing. Adding a # at the end of my number results in a “can not complete call as dialed”. It seems to be more of a firewall issue I am guessing. I have another freepbx distro setup at another location and it doesn’t display this behavior. The outbound routes for these two PBXs are identical (same dial patterns are used in the outbound routes). I am using the same linphone on my laptop to test both setups. Both setups are offsite (neither are local to my laptop’s network). The only real difference I can think of are different firewalls in front of the PBXs. Perhaps some sort of difference in the firewall configurations. Can anyone think of something along those lines that would explain the delay in using one PBX vs the other. It is the same soft IP phone that I am testing with, so I don’t think it can be the phone’s “inefficient dialstring”…
No, firewalls are NOT your problem and guessing is a complete waste of time with computers, they do what you tell them, which is not necessarily what you want them to do ;-), don’t waste your time there, you said the call is completed after a 5 seconds delay, that is just your asterisk server waiting for a signal that you have completed your dial or an unambiguous number, you either don’t have a complete and “closed” dialstring on your phone (which will send that number) or you don’t have a “closed” (unambiguous) outbound route on your server or just that “voipmstrunksip” is not doing a good job. . . That’s all you need to fix.
It seems to be environment specific. I just did two clean installs of the freepbx distro at both sites. I set up an identical iax2 trunk and outbound route on each one. I then used the same ip phone to test both setups. I get the 5 second delay on one. No delay on the other.
Frustrating.
It’s not the end of the world, but it’s just something that I’d really like to know what the cause is. The calls do go through and call quality is excellent. Just that 5 second delay…
Perhapsyou might want to check that your nameservice (by whatever means) to the other site is properly setup, (or just use ip addresses instead of names) .