Hangup detection issues on chans SIP, LOCAL, and ZAP

Still can’t get asterisk to work right. If you’re familiar with my posts, I have two Asterisk Servers in production. I refer to them as my poorly behaved server, and my relatively well behaved server.

The poorly behaved server still has issues with incoming ZAP calls being dropped. I haven’t been able to catch it in the act yet, as it only happens every few weeks, but I did notice this:

SIP/Aron-082ee990 (None) Up AppDial((Outgoing Line)) SIP/Aron-b61b6360 SIP/Aron@park-dial:1 Up Dial(SIP/Aron|30|t) SIP/Maria-b61bbe50 702@outbound:1 Up ParkedCall(702) SIP/Carmen-b61a9b88 s@outbound:1 Up Parked Call() SIP/Aron-082b0e00 (None) Up AppDial((Outgoing Line)) SIP/Aron-b6194948 SIP/Aron@park-dial:1 Up Dial(SIP/Aron|30|t) 6 active channels 3 active calls

Perhaps this has something to with it?

Then there is this problem I’ve always had with the other server. Every few days I’ll check in on it, and see this:

juanita*CLI> show channel
channel       channels      channeltypes  channeltype
juanita*CLI> show channels
Channel              Location             State   Application(Data)
SIP/Maria-0089bae0   (None)               Up      Bridged Call(SIP/Steve-af91f09
SIP/Steve-af91f090   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/Maria-0096da60   (None)               Up      Bridged Call(SIP/Steve-af9b378
SIP/Steve-af9b3780   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/Maria-008a61a0   (None)               Up      Bridged Call(SIP/Steve-af992c8
SIP/Steve-af992c80   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/Maria-0097c280   (None)               Up      Bridged Call(SIP/Steve-af9159d
SIP/Steve-af9159d0   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/steve-007e4e40   (None)               Up      Bridged Call(SIP/Maria-af95006
SIP/Maria-af950060   s@macro-internal_ext Up      Dial(sip/steve|20|tT)
SIP/steve-0086f200   (None)               Up      Bridged Call(SIP/Maria-af936cf
SIP/Maria-af936cf0   s@macro-internal_ext Up      Dial(sip/steve|20|tT)
SIP/Maria-008fccf0   (None)               Up      Bridged Call(SIP/Steve-af90730
SIP/Steve-af907300   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/Maria-00937a50   (None)               Up      Bridged Call(SIP/Steve-af91411
SIP/Steve-af914110   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/steve-008e66a0   (None)               Up      Bridged Call(SIP/Maria-af908d6
SIP/Maria-af908d60   s@macro-internal_ext Up      Dial(sip/steve|20|tT)
SIP/Maria-008e4de0   (None)               Up      Bridged Call(SIP/Steve-af918dc
SIP/Steve-af918dc0   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/Maria-00999010   (None)               Up      Bridged Call(SIP/Steve-af90f1d
SIP/Steve-af90f1d0   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
SIP/steve-00919860   (None)               Up      Bridged Call(SIP/Maria-af95b36
SIP/Maria-af95b360   s@macro-internal_ext Up      Dial(sip/steve|20|tT)
SIP/Maria-008b7450   (None)               Up      Bridged Call(SIP/Steve-af928b4
SIP/Steve-af928b40   s@macro-internal_ext Up      Dial(sip/Maria|20|tT)
26 active channels
13 active calls

While this is an annoyance, this server behaves well most of the year, but for a few episodes quarterly or so.

I could restart them daily and avoid this issue 4 out of 5 times, but that is no way to run a server. I’d like to get to the root of this problem, but so far no luck.

Both servers are nearly identical.

Poorly Behaved Server:
Tyan S2865 MB / Opteron
1GB ECC ram
Gentoo Linux 32-bit
Sangoma A200 card
Snom 360 Phones

Relatively Well Behaved Server:
Poorly Behaved:
Tyan S2865 MB / Opteron
1GB non-ECC ram
Gentoo Linux 32-bit
Sangoma A200 card
Snom 360 and Snom 320 Phones

My main problem with asterisk is its unreliability. Either server will lock up one or all of my few Zap channels every few weeks. They both get to a point where they need a restart. None of my issues are reproducible, and none result in a dump, so the bug tracker is no help.

My hope is that getting insights on the little problems like these will eventually, after years of trying, cause Asterisk to act right.

My goal is very modest: run Asterisk for 30 days with no issues, catastropic or otherwise. Is that realistic?

Any help is much appreciated.

I still have this issue.

Changing PSTN providers from AT&T to Verizon helped with ZAP hangup detection to some extent. I tried that out of desperation despite being assured by everyone including Verizon themselves that it wouldn’t help.

I actually was at the location and witnessed the poorly behaved server mangling an inbound calls. It went like this:

A call comes in and several phones ring.
The call is answered from one of the sip phones
The answering phone plays a dialtone and the other phones continue to ring
Another party attempts to answer the phone, but gets "dead air"
All the phones stop ringing
The original call is “in limbo” between the zap channel and the first failed answer attempt.

Does anyone use FreeSwitch? I don’t know how much longer I can stand this, and was wondering if anyone had good or bad experiences.

what version of asterisk are you running?

Sorry. It’s 1.4.23.1