[asterisk-users] Problems with chan-capi and Eicon Diva 4BRI

[Klaus Darilion]

Hi!

This weekend we had a problem with our Asterisk Box which ran flawlessly 
for nearly 4 weeks. The Asterisk server sits between the PSTN and a 
Siemens PBX and bridges 2 BRI lines. No calls, not incoming, not 
outgoing. The admin rebooted the Dell Box and then everything worked 
fine again.

Now, I'm analyzing log files to find the cause. During the Asterisk 
outage the logfiles only show incoming (PSTN->Asterisk->PBX) calls, no 
outgoing. Thus I suspect that the Asterisk<-->PBX link was broken.

In the Asterisk message file I only see "Recovery on timer expiry" 
errors, like below:

Oct 20 17:18:18 VERBOSE[19772] logger.c:   == ISDN2#02: Incoming call 
'347xxxxx' -> '32xx'
Oct 20 17:18:18 VERBOSE[19772] logger.c:     -- ISDN2#02: Updated 
channel name: CAPI/ISDN2/32xx-8ab6
Oct 20 17:18:18 VERBOSE[2663] logger.c:     -- Executing 
Dial("CAPI/ISDN2/32xx-8ab6", "CAPI/g2//b|90") in new stack
Oct 20 17:18:18 VERBOSE[2663] logger.c:     -- Called g2//b
Oct 20 17:18:19 VERBOSE[19772] logger.c:     -- ISDN2#02: Updated 
channel name: CAPI/ISDN2/32xx11-8ab8
Oct 20 17:18:19 VERBOSE[2663] logger.c:     -- ISDN4#02: Updated channel 
name: CAPI/ISDN4/1-8ab9
Oct 20 17:18:19 VERBOSE[2663] logger.c:     -- ISDN4#02: Updated channel 
name: CAPI/ISDN4/11-8aba
Oct 20 17:18:26 VERBOSE[19772] logger.c:        > ISDN4#02: CAPI INFO 
0x34e6: Recovery on timer expiry
Oct 20 17:18:26 VERBOSE[2663] logger.c:   == ISDN4#02: CAPI Hangingup 
for PLCI=0x104 in state 4
Oct 20 17:18:26 VERBOSE[2663] logger.c:   == Everyone is busy/congested 
at this time (1:0/0/1)
Oct 20 17:18:26 VERBOSE[2663] logger.c:     -- Executing 
Hangup("CAPI/ISDN2/32xx11-8ab8", "") in new stack
Oct 20 17:18:26 VERBOSE[2663] logger.c:   == Spawn extension (frompstn, 
32xx, 2) exited non-zero on 'CAPI/ISDN2/32xx11-8ab8'
Oct 20 17:18:26 VERBOSE[2663] logger.c:   == ISDN2#02: CAPI Hangingup 
for PLCI=0x202 in state 7
Oct 20 17:18:26 VERBOSE[19772] logger.c:        > ISDN2#02: CAPI INFO 
0x34e6: Recovery on timer expiry


What does the timer expiry exactly mean? Was it a Layer2 or Layer 3 
problem? How can I find out more or how can I activate more BRI 
debugging for the case it happens again?

Are there any known problems? We are using:
Asterisk 1.2.12.1
chan_capi-0.7.0
divas4linux-melware-3.0.3-106.650-1
Diva Server 4BRI-8M 2.0 PCI

Thanks
Klaus