<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/3379/">https://reviewboard.asterisk.org/r/3379/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<p style="margin-top: 0;">On March 27th, 2014, 5:14 p.m. CDT, <b>Mark Michelson</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<table width="100%" border="0" bgcolor="white" style="border: 1px solid #C0C0C0; border-collapse: collapse; margin: 2px padding: 2px;">
<thead>
<tr>
<th colspan="4" bgcolor="#F0F0F0" style="border-bottom: 1px solid #C0C0C0; font-size: 9pt; padding: 4px 8px; text-align: left;">
<a href="https://reviewboard.asterisk.org/r/3379/diff/4/?file=56650#file56650line502" style="color: black; font-weight: bold; text-decoration: underline;">/branches/12/res/ari/resource_bridges.c</a>
<span style="font-weight: normal;">
(Diff revision 4)
</span>
</th>
</tr>
</thead>
<tbody>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">469</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="n">ao2_lock</span><span class="p">(</span><span class="n">control</span><span class="p">);</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">470</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="k">if</span> <span class="p">(</span><span class="n">stasis_app_control_is_done</span><span class="p">(</span><span class="n">control</span><span class="p">))</span> <span class="p">{</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">471</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="tb"> </span><span class="n">ast_ari_response_error</span><span class="p">(</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">472</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="tb"> </span><span class="tb"> </span><span class="n">response</span><span class="p">,</span> <span class="mi">500</span><span class="p">,</span> <span class="s">"Internal Error"</span><span class="p">,</span> <span class="s">"Control is about to die"</span><span class="p">);</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">473</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="tb"> </span><span class="n">ao2_unlock</span><span class="p">(</span><span class="n">control</span><span class="p">);</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">474</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="tb"> </span><span class="k">return</span><span class="p">;</span></pre></td>
</tr>
<tr>
<th bgcolor="#b1ebb0" style="border-right: 1px solid #C0C0C0;" align="right"><font size="2"></font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "></pre></td>
<th bgcolor="#b1ebb0" style="border-left: 1px solid #C0C0C0; border-right: 1px solid #C0C0C0;" align="right"><font size="2">475</font></th>
<td bgcolor="#c5ffc4" width="50%"><pre style="font-size: 8pt; line-height: 140%; margin: 0; "><span class="tb"> </span><span class="p">}</span></pre></td>
</tr>
</tbody>
</table>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">A few things here.
First and foremost, this isn't guaranteed to work. Consider that stasis_app_control_execute_until_exhausted() is running in thread 1 and ari_bridges_play_found() is running in thread 2.
Thread 1 calls control_dispatch_all(), which returns 0.
Thread 2 acquires the control lock.
Thread 1 attempts to acquire the control lock but it cannot since thread 2 has the control lock held.
Thread 2 checks if the control is done. It is not, so it decides to attempt the playback.
Thread 2 unlocks the control lock.
Thread 1 locks the control lock, sees that the command queue was exhausted, marks the control as done, and breaks from its loop.
The result is that the queued playfile from thread 2 does not execute because thread 1 has already decided that its command queue is exhausted and has broken free from its loop.
I think the best way to get around this is going to be to alter stasis_app_control_execute_until_exhausted() so that after it acquires the control lock, it checks the container count of the control's command_queue to see if something new has been added. If so, then continue instead of breaking.
The other problem here is that I don't think that if the control is determined to be dying that a 500 response should be returned. Instead, you should fall back to the ari_bridges_play_new() behavior.</pre>
</blockquote>
<p>On March 28th, 2014, 1:29 p.m. CDT, <b>Jonathan Rose</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;">Ah, this is correct. I should have locked prior to setting the control count.</pre>
</blockquote>
</blockquote>
<pre style="margin-left: 1em; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">Oh, that didn't work. Yeah, I had to do a countainer count check afterwards instead otherwise I'd be stuck waiting for commands to finish before I could queue more and that's wrong.</pre>
<br />
<p>- Jonathan</p>
<br />
<p>On March 25th, 2014, 5:14 p.m. CDT, Jonathan Rose wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" style="background-image: url('https://reviewboard.asterisk.org/static/rb/images/review_request_box_top_bg.ab6f3b1072c9.png'); background-position: left top; background-repeat: repeat-x; border: 1px black solid;">
<tr>
<td>
<div>Review request for Asterisk Developers.</div>
<div>By Jonathan Rose.</div>
<p style="color: grey;"><i>Updated March 25, 2014, 5:14 p.m.</i></p>
<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/ASTERISK-22677">ASTERISK-22677</a>
</div>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
Asterisk
</div>
<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;">Previously, if you played an audio file and then played another before the first finished, the second audio file would start playing immediately as it was called overlapping the previous sound. Apparently people don't like that. This patch changes that behavior so that the sound will be queued at the end of any existing controls if they are running.</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 for playback channel wrapper leaks, tested to make sure control objects were being destroyed when they fell out of use. Tested playing of a single file. Tested playing of multiple files in a row. Tested playing of multiple files in a row and then after a sequence finished, playing additional files so that new channels would have to be created. Tested playing sounds right as other sounds were concluding. I wasn't able to break it (although I wouldn't be surprised if there is a possible condition where you can grab a control as it is finishing up its queue and then attempting to add a sound to a finished queue causing the playback to fail. I don't think this would break things in a profound way, it just might possibly make one sound fail to queue under extremely unlikely conditions).</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>/branches/12/rest-api/api-docs/bridges.json <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/stasis/control.c <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/stasis/control.h <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/res_stasis_playback.c <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/res_stasis.c <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/res_ari_bridges.c <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/ari/resource_bridges.c <span style="color: grey">(411187)</span></li>
<li>/branches/12/res/ari/resource_bridges.h <span style="color: grey">(411187)</span></li>
<li>/branches/12/include/asterisk/stasis_app.h <span style="color: grey">(411187)</span></li>
</ul>
<p><a href="https://reviewboard.asterisk.org/r/3379/diff/" style="margin-left: 3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>