|Summary:||ASTERISK-08108: Blind transfers crash Asterisk when verbose = 3|
|Reporter:||John Lange (johnlange)||Labels:|
|Date Opened:||2006-11-10 12:24:43.000-0600||Date Closed:||2006-11-17 09:49:13.000-0600|
|Environment:||Attachments:||( 0) backtrace.txt|
( 1) backtrace02.txt
( 2) consolecapture02.txt
( 3) consolecapture03.txt
|Description:||It seems any blind transfer crashes asterisk when the console is set to verbose 3. This is svn branch 1.4 rev 47345. I have confirmed this happens with both a SPA 942 & SNOM 360 so its not handset specific. I have also confirmed attended transfers are NOT effected.|
When "set verbose 0", it does not crash.
|Comments:||By: Olle Johansson (oej) 2006-11-10 13:00:03.000-0600|
Is this a SIP transfer or a PBX transfer (by using res_features / features.conf transfers) ?
By: John Lange (johnlange) 2006-11-10 13:15:22.000-0600
This is a SIP transfer using the features built into the phone. Also, please remove the "set verbose 3" comment. I have now crashed it with verbose 0 as well so I'm not sure exactly what the determining factor is but I'm trying to lab it so I can provide more information. For the time being lets just say it consistantly crashes on blind transfer regardless of the other settings.
By: John Lange (johnlange) 2006-11-10 16:58:22.000-0600
Console capture 02 and backtrace 02 are from the same segfault. I created this by setting up a minilab on my laptop with 3 phones. I called from phone 1 to phone 2 and blind transferred between phones 2 & 3 several times until it crashed.
__PRETTY_FUNCTION__ = "do_monitor"
ASTERISK-1 0x080fed11 in dummy_start (data=0x81b0360) at utils.c:545
seems to be the common problem. Blind transfer seems to be the fastest way to get it to crash but it also crashes at seemingly random times and the backtrace also mention this function.
By: John Lange (johnlange) 2006-11-10 16:59:18.000-0600
consolecapture03.txt has a very detailed dump of a recent crash also created by doing blind xfers.
By: Kaloyan Kovachev (knk) 2006-11-12 04:34:30.000-0600
it looks to me it is similar to bug 7607. can you please test the patch there if it fixes this one too.
By: John Lange (johnlange) 2006-11-13 12:39:44.000-0600
Ya, this seems like a dup of 7607 (sorry). But note that the patch does not fix the crashes but lets close this and continue in 7607 where I have made some new notes.
By: Serge Vecher (serge-v) 2006-11-17 09:49:13.000-0600
to be continued and fixed in 7607.