chan_sip: fix a deadlock in check_rtp_timeout.

Don't block doing silly deadlock avoidance.  Just return and try again later.
The funciton gets called often enough that it's fine.  Also, this change was
already made in trunk.

(closes issue #18791)
Reported by: irroot
Patches:
      chan_sip.rtptimeout.patch uploaded by irroot (license 52)


git-svn-id: https://origsvn.digium.com/svn/asterisk/branches/1.8@317865 65c4cc65-6c06-0410-ace0-fbb531ad65f3
This commit is contained in:
Russell Bryant
2011-05-06 19:46:49 +00:00
parent 25ccc62ee9
commit d5a0fb899a

View File

@@ -24603,13 +24603,12 @@ static void check_rtp_timeout(struct sip_pvt *dialog, time_t t)
if (!ast_test_flag(&dialog->flags[1], SIP_PAGE2_CALL_ONHOLD) || (ast_rtp_instance_get_hold_timeout(dialog->rtp) && (t > dialog->lastrtprx + ast_rtp_instance_get_hold_timeout(dialog->rtp)))) {
/* Needs a hangup */
if (ast_rtp_instance_get_timeout(dialog->rtp)) {
while (dialog->owner && ast_channel_trylock(dialog->owner)) {
sip_pvt_unlock(dialog);
usleep(1);
sip_pvt_lock(dialog);
}
if (!dialog->owner) {
return; /* channel hangup can occur during deadlock avoidance. */
if (!dialog->owner || ast_channel_trylock(dialog->owner)) {
/*
* Don't block, just try again later.
* If there was no owner, the call is dead already.
*/
return;
}
ast_log(LOG_NOTICE, "Disconnecting call '%s' for lack of RTP activity in %ld seconds\n",
dialog->owner->name, (long) (t - dialog->lastrtprx));