[OpenSER-Devel] [ openser-Bugs-1847821 ] Crash because domain change?
SourceForge.net
noreply at sourceforge.net
Mon Dec 10 13:58:01 UTC 2007
Bugs item #1847821, was opened at 2007-12-10 12:00
Message generated for change (Comment added) made by ibc_sf
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1847821&group_id=139143
Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: modules
Group: ver devel
Status: Open
Resolution: None
Priority: 8
Private: No
Submitted By: Iñaki Baz (ibc_sf)
Assigned to: anca (anca_vamanu)
Summary: Crash because domain change?
Initial Comment:
Hi, I was changing a domain from "sip.aliax.net" to "aliax.net" (SRV type) so I was changing the domain in some OpenSer tables. But during it OpenSer crashes:
Dec 10 12:58:15 [27011] CRITICAL:core:mk_proxy: could not resolve hostname: "aliax.net"
Dec 10 12:58:15 [27011] ERROR:tm:uri2proxy: bad host name in URI <sip:ibc at aliax.net>
Dec 10 12:58:15 [27011] ERROR:tm:uri2su: failed create a dst proxy
Dec 10 12:58:15 [27011] ERROR:tm:t_uac: no socket found
Dec 10 12:58:15 [27011] ERROR:pua:send_publish: in t_request tm module function
Dec 10 12:58:15 [27011] CRITICAL:core:qm_free: freeing already freed pointer, first free: send_publish.c: send_publish(556) - aborting
Dec 10 12:58:15 [27003] INFO:core:handle_sigs: child process 27011 exited by a signal 6
Dec 10 12:58:15 [27003] INFO:core:handle_sigs: core was not generated
Dec 10 12:58:15 [27003] INFO:core:handle_sigs: terminating due to SIGCHLD
Dec 10 12:58:15 [27047] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27044] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27017] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27014] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27008] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27007] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27006] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27005] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27004] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27075] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27072] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27069] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27066] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27063] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27060] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27057] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27054] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27051] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27050] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27041] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27038] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27035] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27032] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27029] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27026] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27023] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27020] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27078] INFO:core:sig_usr: signal 15 received
Dec 10 12:58:15 [27003] INFO:pike:pike_exit: destroying...
Dec 10 12:58:15 [27003] NOTICE:presence:destroy: destroy module ...
Of course this needs more testing and I'll try it but for now it seems a little unexpected to crash because those kinds of changes. Any suggestion or something I could test?
I'm just now in the process of domain transfer so aliax.net is not still propagated so the SRV record doesn't exist for OpenSer. Anyway shouldn't crash because it.
----------------------------------------------------------------------
>Comment By: Iñaki Baz (ibc_sf)
Date: 2007-12-10 13:58
Message:
Logged In: YES
user_id=1844020
Originator: YES
Sorry Bogdan, it occurred because it took long time to me to attach the
log in previous comment and during that time you already did changes.
Sorry.
----------------------------------------------------------------------
Comment By: Bogdan-Andrei Iancu (bogdan_iancu)
Date: 2007-12-10 12:52
Message:
Logged In: YES
user_id=1275325
Originator: NO
please do not change the priority and assignments for the reports! Before
submitting more info on an existing reports, be sure you refresh your web
page.
bogdan
----------------------------------------------------------------------
Comment By: Iñaki Baz (ibc_sf)
Date: 2007-12-10 12:48
Message:
Logged In: YES
user_id=1844020
Originator: YES
ops, sorry, the end of my previous comment just shouldn't exist XD
I set in my resolv.conf a non existent domain (testdns.deb)pointing to my
OpenSer to try if it crashes too, but it doesn't crash (both aliax.net and
testdns.deb exist in "domain" table).
----------------------------------------------------------------------
Comment By: Iñaki Baz (ibc_sf)
Date: 2007-12-10 12:42
Message:
Logged In: YES
user_id=1844020
Originator: YES
More data:
OpenSer crashes during a REGISTER like:
REGISTER sip:ibc at aliax.net
if OpenSer can't resolve that domain (aliax.net). But it doesn't occur
with other not resolvable domains. ¿?¿
So it should be because some table containing domain name in some fields.
I attach a debug.
REGISTER sip:testdns.deb -- desde Twinkle/1.1alpha2 -
212.121.235.18:1031
FROM: sip:ibc at testdns.deb
TO: sip:ibc at testdns.deb
NAT detectado
ERROR: REGISTER en Outbound no permitido
branch_route[1]: ru = sip:testdns.deb
fr_inv_timer = <null> seconds
bflag(6) = 1
Dec 10 13:09:44 [27558] CRITICAL:core:mk_proxy: could not resolve
hostname: "testdns.deb"
Dec 10 13:09:44 [27558] ERROR:tm:uri2proxy: bad host name in URI
<sip:testdns.deb>
Dec 10 13:09:44 [27558] ERROR:tm:t_forward_nonack: failure to add
branches
File Added: openser_crash_register_aliax.net
----------------------------------------------------------------------
Comment By: Bogdan-Andrei Iancu (bogdan_iancu)
Date: 2007-12-10 12:34
Message:
Logged In: YES
user_id=1275325
Originator: NO
looks like the PUA module does not properly handle some error cases.
bogdan
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=743020&aid=1847821&group_id=139143
More information about the Devel
mailing list