I just upgraded one of our Switchvox AA65 PBX’s to ver 5.1.1. Everything worked fine until then. Now all internal calls hang for 12-15 seconds before ringing the phone. It is just the phones, IVR, virtual extensions, Call Queues and Voice Mail pickup right away. If they call an extension on another PBX there is no delay. In the case of the SIP phones if they are already on a call, the ring does not delay on a new call. Only does it from a cold start. I have 3 other PBX’s, one is a AA65, running version 5.1.1 and they do not have this problem.
Can you describe your network solution (routers, switches or etc)? Also how familiar are you with Asterisk? Do you know how to check what is happening in the background (Asterisk CLI and SIP debug) ?
I’m having this same problem (did an upgrade on 10/2 from 4.x to 5.1.1, now experiencing 15ish second call delays, and some internal calls don’t allow the call recipient to speak for 5 seconds or so). I have a ticket in to Digium to assist, but we’re only on Silver support, so I’m not able to get on the phone with them.
Looking at other asterisk installations (like Trixbox), it seems like fax detection, time zone, and caller id all play a part in call delays. Most of the forum posts I’ve seen, though, seem to be more concerned with incoming call delays instead of internal call delays.
I’m somewhat familiar with what’s going on behind the scenes with the cli and all that, but I’m no pro.
Edit: PBX is a Switchvox 305 system.
An update: the internal call delay was resolved by Digium tech support. They had me change my Jabber server hostname from the external IP of the system to “pbx.” That was it; as soon as the network restarted the delay was gone.
Digium had us do the same but not a complete fix only helped.
It turns out that if the Load is in Alarm (exceeded the threshold)[Under Diagnostics/server information] it does not display an error in the banner nor does it show up in the error log. The only way you can find this is by going to the Server Information Page. The average load was running at 8.69 and the threshold is 5.
Why does this not show up in the banner where other message are displayed. Something as important as this you would think should be display in the banner or at the very least the Error Log. This is something that needs to be fixed. The new interface is nice looking but it does nothing for the performance of the device.
To fix this you have physically power cycle the device. I ran a server reboot twice, a software reload 4 times and a software restart twice. None of that helped. After seeing that the load was over the threshold I drove there and power cycled it.