Vicidial 2.12-504a build 150804-1608
3 server cluster (1 DB + 2 dialers)
- 1 dialer is Vicibox 4.0.3
- 1 dialer is CentOS scratch install
- DB is CentOS
The problem I´m having is that randomly, call recordings get 0 length. The table vicidial_recorgins report 0 seconds length, and the files (both .wav and .mp3) are too small in size, less than 1 K.
Agent time reports are correct, so calls that report 1000 secs total, have 0 seconds in recordings.
All campaigns have ALLFORCE recording.
This happens randomly in all the campaigns: most of the recordings are OK and process correctly, but some are just 0 secs long.
I´ve checked Asterisk logs but in these cases, the Monitor extensions gets hung up immediately after starting. It looks like this:
- Code: Select all
VERBOSE[14381] logger.c: [Sep 14 11:12:13] -- Executing [8309@default:1] Answer("Local/58600054@default-4cc5,1", "") in new stack
VERBOSE[14381] logger.c: [Sep 14 11:12:13] -- Executing [8309@default:2] Monitor("Local/58600054@default-4cc5,1", "wav|20150914-111212_6861447967") in new stack
VERBOSE[14381] logger.c: [Sep 14 11:12:13] -- Executing [8309@default:3] Wait("Local/58600054@default-4cc5,1", "3600") in new stack
VERBOSE[14381] logger.c: [Sep 14 11:12:13] == Spawn extension (default, 8309, 3) exited non-zero on 'Local/58600054@default-4cc5,1'
VERBOSE[14381] logger.c: [Sep 14 11:12:13] -- Executing [h@default:1] DeadAGI("Local/58600054@default-4cc5,1", "agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----0---------------") in new stack
VERBOSE[14381] logger.c: [Sep 14 11:12:13] -- AGI Script agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----0--------------- completed, returning 0
The call starts and ends within the same second.
Does anyone have any idea of what could be causing this immediate hang up for the recording?