<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="8" style="border: 1px #c9c399 solid;">
<tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://reviewboard.asterisk.org/r/1286/">https://reviewboard.asterisk.org/r/1286/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<p style="margin-top: 0;">On July 12th, 2011, 1:42 p.m., <b>opticron</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">The added if()s alternate between checking for NULL and using ast_strlen_zero. ast_strlen_zero would probably be more appropriate in all cases.</pre>
</blockquote>
<p>On July 12th, 2011, 5:10 p.m., <b>jamuel</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Testing for ast_strlen_zero in all instances can't be done since mem can be NULL and ast_strlen_zero will segfault if it tests deferencing a structure that is initialized as NULL.</pre>
</blockquote>
<p>On July 12th, 2011, 7:24 p.m., <b>Tilghman Lesher</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Yes, it can be used, and no, it won't segfault. ast_strlen_zero() returns true if the value passed is itself NULL.</pre>
</blockquote>
<p>On July 13th, 2011, 2:46 p.m., <b>jamuel</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">In rqm_exec (line 5735), tried testing with ast_strlen_zero(mem->membername) instead of mem->membername == NULL and I get a segfault:
Breakpoint 1, rqm_exec (chan=0xf502d30, data=0xb7395c08 "4152000400,Local/4152000501@from-internal/n") at app_queue.c:5735
5735 if (ast_strlen_zero(mem->membername) || !log_membername_as_agent) {
(gdb) step
53 if (!s || (*s == '\0')) {
(gdb) p mem
$1 = (struct member *) 0x0
(gdb) step
Program received signal SIGSEGV, Segmentation fault.
0x00c72627 in rqm_exec (chan=0xf502d30, data=0xb7395c08 "4152000400,Local/4152000501@from-internal/n") at /usr/local/src/asterisk-TRUNK/include/asterisk/strings.h:53
53 if (!s || (*s == '\0')) {
I must be doing something stupid. Any help would be greatly appreciated.</pre>
</blockquote>
<p>On July 14th, 2011, 9:10 a.m., <b>opticron</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">The danger of segfault is with mem being NULL, not mem->membername. If you test log_membername_as_agent first, it's less likely to happen, but still possible. You'll want to ensure that mem is not NULL before dereferencing it.</pre>
</blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">So without complicating the test any further I'll leave the test as mem == NULL--this should always be safe and eliminate the potential for segfault corner-cases.</pre>
<br />
<p>- jamuel</p>
<br />
<p>On July 12th, 2011, 5:12 p.m., jamuel wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" style="background-image: url('https://reviewboard.asterisk.org/media/rb/images/review_request_box_top_bg.png'); background-position: left top; background-repeat: repeat-x; border: 1px black solid;">
<tr>
<td>
<div>Review request for Asterisk Developers, Tilghman Lesher, leifmadsen, opticron, and irroot.</div>
<div>By jamuel.</div>
<p style="color: grey;"><i>Updated July 12, 2011, 5:12 p.m.</i></p>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: 1px solid #b8b5a0">
<tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">app_queue logs the events ADDMEMBER and REMOVEMEMBER with the agent field set to the interface value rather than the membername value when a member is added with a state_interface value set. However all other member related queue events are logged with the membername when a state_interface is set. It would seem that these ADD/REMOVE MEMBER events should log the agent field in the same manner.
This patch addresses the original issue reported in https://issues.asterisk.org/jira/browse/ASTERISK-14769 by providing a queues.conf general configuration option log_membername_as-agent. Setting log_membername_as_agent = yes will cause the membername to be logged in the agent field for ADDMEMBER and REMOVEMEMBER queue events if a state_interface has been set.
The default value (log_membername_as_agent = no) does not alter the existing logging for the ADDMEMBER and REMOVEMEMBER events to maintain existing (buggy) behavior for the sake of backwards compatibility. </pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Testing </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: 1px solid #b8b5a0">
<tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Tested add, remove, pause, unpause via AMI, CLI, and Dialplan apps and see the correct log output to queue_log. Also tested with log_membername_as_agent omitted to insure backward compatibility for those queues.conf that might not be updated for this new option.
Sample test plan (add and remove)
---------------------------------
set log_membername_as_agent = yes in queues.conf
AMI:
====
Action: QueueAdd
Queue: 4152000400
Interface: Local/4152000501@from-internal/n
Penalty: 0
Paused: 0
MemberName: Edward Frank
Action: QueueRemove
Queue: 4152000400
Interface: Local/4152000501@from-internal/n
Action: QueueAdd
Queue: 4152000400
Interface: Local/4152000501@from-internal/n
Action: QueueRemove
Queue: 4152000400
Interface: Local/4152000501@from-internal/n
CLI:
====
> queue add member Local/4152000501@from-internal/n to 4152000400 penalty 0 as "Edward Frank" state_interface Sip/SOFTJPS
> queue remove member Local/4152000501@from-internal/n from 4152000400
> queue add member Local/4152000501@from-internal/n to 4152000400
> queue remove member Local/4152000501@from-internal/n from 4152000400
Dialplan:
=========
[from-test]
exten => 1234,1,AddQueueMember(4152000400,Local/4152000501@from-internal/n,0,,Edward Davis,Sip/SOFTJPS)
exten => 1234,n,RemoveQueueMember(4152000400,Local/4152000501@from-internal/n)
exten => 1234,n,Hangup()
exten => 5678,1,AddQueueMember(4152000400,Local/4152000501@from-internal/n,0)
exten => 5678,n,RemoveQueueMember(4152000400,Local/4152000501@from-internal/n)
exten => 5678,n,Hangup()
Rinse and repeat with log_membername_as_agent = no
Repeat with log_membername_as_agent omitted.
</pre>
</td>
</tr>
</table>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Bugs: </b>
<a href="https://issues.asterisk.org/jira/browse/https://issues.asterisk.org/jira/browse/ASTERISK-14769">https://issues.asterisk.org/jira/browse/ASTERISK-14769</a>
</div>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>http://svn.asterisk.org/svn/asterisk/trunk/apps/app_queue.c <span style="color: grey">(327949)</span></li>
<li>http://svn.asterisk.org/svn/asterisk/trunk/configs/queues.conf.sample <span style="color: grey">(327949)</span></li>
</ul>
<p><a href="https://reviewboard.asterisk.org/r/1286/diff/" style="margin-left: 3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>