I recently upgraded from an old TrixBox to a new AsteriskNow after a system crash. I tried to upgrade to TrixBox along the way, but could not get the new version to work in my environment, so after speaking with Digium (my card provider) I went to AsteriskNow - which is working, sort of…
Problem: Everything appears to work correctly when using ring stragety “ringfirstavailable” but swiching to “ringall” causes inbound calls to drive cpu to 100% in dialparties.agi without every deciding what to do. Debugging seems to be difficult as the problem does not reproduce reliably - although when it starts to happen it keeps happening until at least one reboot - althoug it does not always correct after a reboot.
Ringfirstavailable is not a good solution in this environment as there is some distance between phones and not everyone is present at all times.
Environment: POTS lines -> Digium TDM2400P with 6 FXO ports on two daughter cards -> DAHDI channels -> Ring Groups -> SIP phones (Polycom, Cisco & Granstream) and SIP softphone (eyeBeam). On no answer, then use backup ring group that adds outbound SIP to answering service.
System was installed from AsteriskNow 1.5.0 CD then all updates applied.
I have not applied the recent gnutls 1.4.1-3 or kernel & headers 2.6.18-128.7.1.el5 or kmod-dahdi-linux 220.127.116.11-1_centos18.104.22.168_128.7.1.el5 as I have been WAY out of town since before they were available and do not want to zap the production system when I cannot be there to repair it.
Not sure what else anyone needs to know to help with this - just let me know and I will get the info.
Please help. Any information on how to proceed to isolate the issue or to correct it would be helpful.