[asterisk-bugs] [Asterisk 0015894]: SIP register via tls causes lock on sip reload
Asterisk Bug Tracker
noreply at bugs.digium.com
Thu Oct 22 15:00:25 CDT 2009
The following issue has been ASSIGNED.
======================================================================
https://issues.asterisk.org/view.php?id=15894
======================================================================
Reported By: dvossel
Assigned To: dvossel
======================================================================
Project: Asterisk
Issue ID: 15894
Category: Channels/chan_sip/TCP-TLS
Reproducibility: always
Severity: major
Priority: normal
Status: assigned
Asterisk Version: SVN
JIRA:
Regression: No
Reviewboard Link: https://reviewboard.asterisk.org/r/380/
SVN Branch (only for SVN checkouts, not tarball releases): trunk
SVN Revision (number only!): 218430
Request Review:
======================================================================
Date Submitted: 2009-09-14 17:47 CDT
Last Modified: 2009-10-22 15:00 CDT
======================================================================
Summary: SIP register via tls causes lock on sip reload
Description:
I ran into this problem because of another issue Asterisk has with TLS.
register => tls://user:1234@mydomain
Given the register line above in sip.conf, I'd expect the register request
to be send via port 5061. Instead it defaults to port 5060 which is
incorrect for TLS.
----------------------------------------------------------------
*CLI> sip show registry
Host dnsmgr Username Refresh State
Reg.Time
mydomain:5060 N user 120 Request Sent
----------------------------------------------------------------
Now here's the odd part. I decided to go into my sip.conf and explicitly
set the register line to have a port, and when I issued a sip reload
everything locked up, even the CLI. I was able to reproduce the steps and
got the same results everytime.
It seems that this is somehow caused by the receiving end binding to port
5061 and me sending the registration using port 5060. If I bind the
receiving end to 5060 this works even though that is not the correct port
for tls. I tried another test where I pointed the register at a fake
domain, but nothing locked up.
======================================================================
Relationships ID Summary
----------------------------------------------------------------------
has duplicate 0015343 [patch] lock in sip_tcp_helper_thread
related to 0013136 [patch] sip peer qualified failed, aste...
======================================================================
----------------------------------------------------------------------
(0112629) svnbot (reporter) - 2009-10-22 15:00
https://issues.asterisk.org/view.php?id=15894#c112629
----------------------------------------------------------------------
Repository: asterisk
Revision: 225445
U trunk/apps/app_externalivr.c
U trunk/channels/chan_sip.c
U trunk/include/asterisk/tcptls.h
U trunk/main/tcptls.c
------------------------------------------------------------------------
r225445 | dvossel | 2009-10-22 15:00:19 -0500 (Thu, 22 Oct 2009) | 50
lines
SIP TCP/TLS: move client connection setup/write into tcp helper thread,
various related locking/memory fixes.
What this patch fixes
1.Moves sip TCP/TLS connection setup into the TCP helper thread:
Connection setup takes awhile and before this it was being
done while holding the monitor lock.
2.Moves TCP/TLS writing to the TCP helper thread: Through the
use of a packet queue and an alert pipe, the TCP helper thread
can now be woken up to write data as well as read data.
3.Locking error: sip_xmit returned an XMIT_ERROR without giving
up the tcptls_session lock. This lock has been completely removed
from sip_xmit and placed in the new sip_tcptls_write() function.
4.Memory leak: When creating a tcptls_client the tls_cfg was alloced
but never freed unless the tcptls_session failed to start. Now the
session_args for a sip client are an ao2 object which frees the
tls_cfg on destruction.
5.Pointer to stack variable: During sip_prepare_socket the creation
of a client's ast_tcptls_session_args was done on the stack and
stored as a pointer in the newly created tcptls_session. Depending
on the events that followed, there was a slight possibility that
pointer could have been accessed after the stack returned. Given
the new changes, it is always accessed after the stack returns
which is why I found it.
Notable code changes
1.I broke tcptls.c's ast_tcptls_client_start() function into two
functions. One for creating and allocating the new tcptls_session,
and a separate one for starting and handling the new connection.
This allowed me to create the tcptls_session, launch the helper
thread, and then establish the connection within the helper thread.
2.Writes to a tcptls_session are now done within the helper thread.
This is done by using an alert pipe to wake up the thread if new
data needs to be sent. The thread's sip_threadinfo object contains
the alert pipe as well as the packet queue.
3.Since the threadinfo object contains the alert pipe, it must now be
accessed outside of the helper thread for every write (queuing of a
packet). For easy lookup, I moved the threadinfo objects from a
linked list to an ao2_container.
(closes issue https://issues.asterisk.org/view.php?id=13136)
Reported by: pabelanger
Tested by: dvossel, whys
(closes issue https://issues.asterisk.org/view.php?id=15894)
Reported by: dvossel
Tested by: dvossel
Review: https://reviewboard.asterisk.org/r/380/
------------------------------------------------------------------------
http://svn.digium.com/view/asterisk?view=rev&revision=225445
Issue History
Date Modified Username Field Change
======================================================================
2009-10-22 15:00 svnbot Checkin
2009-10-22 15:00 svnbot Note Added: 0112629
2009-10-22 15:00 svnbot Status ready for review =>
assigned
======================================================================
More information about the asterisk-bugs
mailing list