Hi everybody,
I have a Fedora Core 4 x86 32 bit install, which I recently
upgraded from asterisk 1.2 to the office 1.4.4 tarball.
In the process of doing that I had to upgrade some
autoconf/automake stuff, but it worked fine, and my new asterisk
works fine.
Except that anytime I receive a fax with spandsp and app_rxfax,
asterisk seg faults.
I have applied the spandsp patch of course, and I used the newer
app_rxfax.c and app_txfax.c from soft-switch.org for 1.4. I have
tried numerous versions of spandsp in turn, recompiling the rxfax
and txfax application after installing each one. In particular, I
am still getting this when using the latest spandsp snapshot from
June 8th.
I my searching, I found a few other people who mentioned the same
problem, but they either didn't say if they solved it or were vague
about how they did.
My verison of libtiff is 3.7.1.
One of the crashes printed out this stack trace information,
although that doesn't come out most of the time:
XXX
-- Executing [EMAIL PROTECTED]:2] Set(Zap/1-1, [EMAIL PROTECTED]) in
new stack
-- Executing [EMAIL PROTECTED]:3] RxFAX(Zap/1-1,
/var/spool/asterisk/fax/1181537898.0.tif|debug) in new stack
linux40*CLI *** glibc detected *** /usr/sbin/asterisk: free(): invalid next
size (normal): 0x09d5a908 ***
=== Backtrace: =
/lib/libc.so.6[0x5471e0]
/lib/libc.so.6(__libc_free+0x77)[0x54772b]
/usr/lib/asterisk/modules/app_rxfax.so[0xebbc41]
/usr/sbin/asterisk[0x80c3cc8]
/usr/lib/asterisk/modules/app_macro.so[0xa330ea]
/usr/sbin/asterisk[0x80c3cc8]
/usr/sbin/asterisk[0x80c5002]
/usr/sbin/asterisk[0x80c5d3e]
/usr/sbin/asterisk[0x80f1f99]
/lib/libpthread.so.0[0x656bd4]
/lib/libc.so.6(__clone+0x5e)[0x5ae4fe]
=== Memory map:
00111000-00118000 r-xp 03:01 5330798
/usr/lib/asterisk/modules/res_musiconhold.so
00118000-00119000 rwxp 7000 03:01 5330798
/usr/lib/asterisk/modules/res_musiconhold.so
00119000-0011f000 r-xp 03:01 5330792
/usr/lib/asterisk/modules/res_config_pgsql.so
0011f000-0012 rwxp 5000 03:01 5330792
/usr/lib/asterisk/modules/res_config_pgsql.so
0012-00218000 r-xp 03:01 196501
/opt/lumenvox/engine_6.5/lib/libcrypto.so.0.9.7f
00218000-0022a000 rwxp 000f8000 03:01 196501
/opt/lumenvox/engine_6.5/lib/libcrypto.so.0.9.7f
0022a000-0022d000 rwxp 0022a000 00:00 0
X
The memory map continues on for several pages but I can supply it
if anyone thinks it would be useful.
Most of the time the *CLI prompt just shows asterisk
disconnecting, and the /var/log/asterisk/full just shows RxFax as
the last thing ran, and the message of asterisk starting up again,
like this:
X
[Jun 10 18:26:12] VERBOSE[13094] logger.c: -- Executing [EMAIL
PROTECTED]:2] Set(Za
p/1-1, [EMAIL PROTECTED]) in new stack
[Jun 10 18:26:12] DEBUG[13094] app_macro.c: Executed application: Set
[Jun 10 18:26:12] VERBOSE[13094] logger.c: -- Executing [EMAIL
PROTECTED]:3] RxFAX(
Zap/1-1, /var/spool/asterisk/fax/1181517967.7.tif|debug) in new stack
[Jun 10 18:26:17] VERBOSE[13163] logger.c: Asterisk Event Logger Started
/var/log/asterisk
/event_log
[Jun 10 18:26:17] VERBOSE[13163] logger.c: Asterisk Dynamic Loader Starting:
[Jun 10 18:26:17] VERBOSE[13163] logger.c: == Parsing
'/etc/asterisk/modules.conf': [Jun
10 18:26:17] VERBOSE[13163] logger.c: Found
X
Is anyone else out there seeing this ?
Does anyone have any suggestions, even if it is only how to get more
debuging information out ?
From the reading and searching I have been doing the last several
hours, it appears that the best thing to do in the long run might be
to install iaxmodem and HylaFax. However, right now I have my
asterisk invoking some custom scripts and uploading the faxes into a
database based on DID and CallerID numbers, and I would much rather
get this working now this way instead of have to learn how to
integrate the same thing with HylaFax.
Thanks in advance,
--Rob
--
http://rgr.freeshell.org/
___
--Bandwidth and Colocation provided by Easynews.com --
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
