Unexpected Hangup (hangup cause 31)

I have an interesting issue that I cannot seem to find reference to the forums.

We recently installed an Asterisk system and we are using a PRI. On some calls, primarily to cell phone subscribers, we get a hangup request about the time the call should go to callers voice mail. Dialing the same number on a local POTS line goes all the way through to voicemail, but on the PRI it disconnects right at about 20 seconds (about the time voicemail transfer kicks in).

Debug output at time of disconnect is shown below. I looked up the mapping for Cause Code 31, which is rather generic and provided no clues. I could not find any information on the PRI_EVENT_HANGUP_ACK(9).

Anyone have any pointers on how to track this down?

[Jan 8 15:28:29] PRI Span: 1 < TEI: 0 State 7(Multi-frame established)
[Jan 8 15:28:29] PRI Span: 1 < V(A)=4, V(S)=4, V®=7
[Jan 8 15:28:29] PRI Span: 1 < K=7, RC=0, l3_initiated=0, reject_except=0, ack_pend=0
[Jan 8 15:28:29] PRI Span: 1 < T200_id=0, N200=3, T203_id=8192
[Jan 8 15:28:29] PRI Span: 1 < [ 02 01 0e 08 08 02 80 02 5a ]
[Jan 8 15:28:29] PRI Span: 1 < Informational frame:
[Jan 8 15:28:29] PRI Span: 1 < SAPI: 00 C/R: 1 EA: 0
[Jan 8 15:28:29] PRI Span: 1 < TEI: 000 EA: 1
[Jan 8 15:28:29] PRI Span: 1 < N(S): 007 0: 0
[Jan 8 15:28:29] PRI Span: 1 < N®: 004 P: 0
[Jan 8 15:28:29] PRI Span: 1 < 5 bytes of data
[Jan 8 15:28:29] PRI Span: 1 < Protocol Discriminator: Q.931 (8) len=5
[Jan 8 15:28:29] PRI Span: 1 < TEI=0 Call Ref: len= 2 (reference 2/0x2) (Sent to originator)
[Jan 8 15:28:29] PRI Span: 1 < Message Type: RELEASE COMPLETE (90)
[Jan 8 15:28:29] PRI Span: 1 – Got ACK for N(S)=4 to (but not including) N(S)=4
[Jan 8 15:28:29] PRI Span: 1 – T200 requested to stop when not started
[Jan 8 15:28:29] PRI Span: 1 T203 requested to start without stopping first
[Jan 8 15:28:29] PRI Span: 1 – Starting T203 timer
[Jan 8 15:28:29] PRI Span: 1 Received message for call 0x7f5dc400b050 on link 0x2469730 TEI/SAPI 0/0
[Jan 8 15:28:29] PRI Span: 1 q931.c:9204 post_handle_q931_message: Call 32770 enters state 0 (Null). Hold state: Idle
[Jan 8 15:28:29] PRI Span: 1 q931.c:7332 q931_hangup: Hangup other cref:32770
[Jan 8 15:28:29] PRI Span: 1 q931.c:7089 __q931_hangup: ourstate Null, peerstate Null, hold-state Idle
[Jan 8 15:28:29] PRI Span: 1 Destroying call 0x7f5dc400b050, ourstate Null, peerstate Null, hold-state Idle
[Jan 8 15:28:29] PRI Span: 1
[Jan 8 15:28:29] PRI Span: 1 > TEI: 0 State 7(Multi-frame established)
[Jan 8 15:28:29] PRI Span: 1 > V(A)=4, V(S)=4, V®=8
[Jan 8 15:28:29] PRI Span: 1 > K=7, RC=0, l3_initiated=0, reject_except=0, ack_pend=0
[Jan 8 15:28:29] PRI Span: 1 > T200_id=0, N200=3, T203_id=8192
[Jan 8 15:28:29] PRI Span: 1 > [ 02 01 01 10 ]
[Jan 8 15:28:29] PRI Span: 1 > Supervisory frame:
[Jan 8 15:28:29] PRI Span: 1 > SAPI: 00 C/R: 1 EA: 0
[Jan 8 15:28:29] PRI Span: 1 > TEI: 000 EA: 1
[Jan 8 15:28:29] PRI Span: 1 > Zero: 0 S: 0 01: 1 [ RR (receive ready) ]
[Jan 8 15:28:29] PRI Span: 1 > N®: 008 P/F: 0
[Jan 8 15:28:29] PRI Span: 1 > 0 bytes of data
[Jan 8 15:28:29] PRI Span: 1 Done handling message for SAPI/TEI=0/0
[Jan 8 15:28:29] Span 1: Processing event PRI_EVENT_HANGUP_ACK(9)
[Jan 8 15:28:29] VERBOSE[10525][C-00000001] sig_pri.c: Span 1: Channel 0/2 got hangup request, cause 31
[Jan 8 15:28:29] VERBOSE[11191] chan_dahdi.c: PRI Span: 1 q931.c:7332 q931_hangup: Hangup other cref:32770
[Jan 8 15:28:29] VERBOSE[11191] chan_dahdi.c: PRI Span: 1 q931.c:7089 __q931_hangup: ourstate Disconnect Indication, peerstate Disconnect Request, hold-state Idle
[Jan 8 15:28:29] VERBOSE[11191] chan_dahdi.c: PRI Span: 1 q931.c:6124 q931_release: Call 32770 enters state 19 (Release Request). Hold state: Idle
[Jan 8 15:28:29] VERBOSE[11191] chan_dahdi.c: PRI Span: 1

Cause 31 is normal clearing. I think you have to look to the network.

Thanks for the reply.

Yes, as I look deeper at the debug output above, it does appear the PRI is reporting that it is receiving a hangup message.