| Summary: | ASTERISK-01747: MOH. server freezes when loading 10 chars after loading music_onhold | ||
| Reporter: | zoa (zoa) | Labels: | |
| Date Opened: | 2004-06-02 17:50:30 | Date Closed: | 2011-06-07 14:01:00 |
| Priority: | Blocker | Regression? | No |
| Status: | Closed/Complete | Components: | Core/General |
| Versions: | Frequency of Occurrence | ||
| Related Issues: | |||
| Environment: | Attachments: | ||
| Description: | I just finished upgrading the asterisk on my server. zaptel is loaded, wcfxo is loaded. i do a ztcfg asterisk -vvvvvvvvvvgc == Registered channel type 'Modem' (Generic Voice Modem Channel Driver) [res_musiconhold.so] => (Music On Hold Resource) == Parsing '/etc/asterisk/musiconhold.conf': Found == Registered application 'MusicOnHold' == Registered application 'WaitMusicOnHold' == Registered application 'SetMusicOnHold' [res_adsi.so] *server hangs completely* ****** ADDITIONAL INFORMATION ****** when i noload moh, it succeeds to go further but then stops at the chan_sip line, because chan_sip needs moh. I'm running 2.4.26 on debian stable gcc 2.95. I've seen such lockups before and they went away when upgrading to 2.6 kernel series. (which is not an option here at home.) Any suggestions on debugging this puppy ? only 2 more modules are loaded, ipsec and a networkcard module, i also tried without the networkcard to see if a change in the order would make a difference, it didn't. the server is running yesterdays cvs. | ||
| Comments: | By: Mark Spencer (markster) 2004-06-02 18:25:30 how about finding me on IRC so I can take a look and see what's going on. Thanks. (irc.freenode.net, #asterisk, kram) By: zoa (zoa) 2004-06-02 19:07:13 Of course, when i wanted to show kram, it worked normally again... (although the install wasnt exactly the same any more). I'm pretty sure this is not a bug in asterisk itself, i think it has something to do with the gcc2.95 version. I remember vagely that i had a similar issue on a dual xeon a long time ago, and then i worked around it by upgrading to kernel 2.6.x. At that time, it also seemed to depend more on module sizes, or compile dates than on the actual source itself. i'll close this bug report and leave this as a reference to people that might experience the same thing someday. | ||