<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
On 08/23/2011 08:38 PM, Frédéric Trottier-Hébert wrote:
<blockquote
cite="mid:AFAC5355-84A9-4307-A5F8-A7A2419B87A9@erlang-solutions.com"
type="cite">
<div>
<div>On 2011-08-23, at 03:48 AM, Henrik Nord wrote:</div>
<br class="Apple-interchange-newline">
<blockquote type="cite">
<div>On 08/03/2011 05:07 PM, Frédéric Trottier-Hébert wrote:<br>
<blockquote type="cite">git fetch
<a class="moz-txt-link-abbreviated" href="mailto:git@github.com:ferd/otp.git">git@github.com:ferd/otp.git</a>
fix_supervisor_temporary_restart<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">This patch fixes the issue discussed
at <a moz-do-not-send="true"
href="http://erlang.org/pipermail/erlang-questions/2011-August/060419.html">http://erlang.org/pipermail/erlang-questions/2011-August/060419.html</a>,
namely that in one_for_all or rest_for_one strategies, the
supervisor is still trying to restart temporary children.
Because we don't track their MFAs entirely, this causes
repetitive crashes.<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">This patch fixes the behaviour by
inserting a clause in terminate_children/2-3 (private
function) that will omit temporary children when building
a list of killed processes, to avoid having the supervisor
trying to restart them again.<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">Only supervisors in need of
restarting children used the list, so the change should be
of no impact for the functions that called
terminate_children/2-3 only to kill all children. This
also avoids useless iterations if the filtering step were
to be inserted at any other point in the code.<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">A test suite has been added to
supervisor_SUITE, containing checks for both one_for_all
and rest_for_one. I've only ran the tests for the suite
itself, but all of them pass without a problem.<br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">The commit/diff itself: <a
moz-do-not-send="true"
href="https://github.com/ferd/otp/commit/25aab8efdef5c6898a679705158627cded3f847a">https://github.com/ferd/otp/commit/25aab8efdef5c6898a679705158627cded3f847a</a><br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">--<br>
</blockquote>
<blockquote type="cite">Fred Hébert<br>
</blockquote>
<blockquote type="cite"><a moz-do-not-send="true"
href="http://www.erlang-solutions.com">http://www.erlang-solutions.com</a><br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite"><br>
</blockquote>
<blockquote type="cite">_______________________________________________<br>
</blockquote>
<blockquote type="cite">erlang-patches mailing list<br>
</blockquote>
<blockquote type="cite"><a moz-do-not-send="true"
href="mailto:erlang-patches@erlang.org">erlang-patches@erlang.org</a><br>
</blockquote>
<blockquote type="cite"><a moz-do-not-send="true"
href="http://erlang.org/mailman/listinfo/erlang-patches">http://erlang.org/mailman/listinfo/erlang-patches</a><br>
</blockquote>
<br>
Hello<br>
<br>
Your patch has been reviewed, and we want you to fix the
following for it to be graduated.<br>
<br>
1) I think the test needs to be extended a bit. I can't see
that it checks what happens to the temporary child - i.e.
that it is terminated but not restarted. As far as I can
understand it only checks that the supervisor survives...<br>
<br>
2) There is a minor typo in the description of the test - it
says one_for_rest instead of rest_for_one.<br>
<br>
3) The behaviour must be documented - i.e. it must be
explicitly stated that temporary children under one_for_all
or rest_for_one supervisors will be terminated but not
restarted.<br>
<br>
/siri<br>
<br>
<br>
Thank you.<br>
<br>
-- <br>
/Henrik Nord Erlang/OTP<br>
<br>
</div>
</blockquote>
</div>
<br>
1) Good catch. The broken behaviour would crash the supervisor --
and I tested that this was solved, but not that the child isn't
retarted. This has been fixed.<br>
<br>
2) also fixed<br>
<br>
3) I changed it. I wasn't too sure where to put it, but there was
only one place where temporary children were discussed so I added
it there.<br>
<br>
I've amended the commit: <a moz-do-not-send="true"
href="https://github.com/ferd/otp/commit/4f9b938112a80f1c1e210b1a6af40a42f833cfa2">https://github.com/ferd/otp/commit/4f9b938112a80f1c1e210b1a6af40a42f833cfa2</a><br>
<br>
The fetch command was: git fetch <a class="moz-txt-link-abbreviated" href="mailto:git@github.com:ferd/otp.git">git@github.com:ferd/otp.git</a>
fix_supervisor_temporary_restart<br>
<br>
Thanks for your time,<br>
<br>
--<br>
Fred Hébert<br>
<a moz-do-not-send="true" href="http://www.erlang-solutions.com">http://www.erlang-solutions.com</a>
<pre wrap="">
<fieldset class="mimeAttachmentHeader"></fieldset>
_______________________________________________
erlang-patches mailing list
<a class="moz-txt-link-abbreviated" href="mailto:erlang-patches@erlang.org">erlang-patches@erlang.org</a>
<a class="moz-txt-link-freetext" href="http://erlang.org/mailman/listinfo/erlang-patches">http://erlang.org/mailman/listinfo/erlang-patches</a>
</pre>
</blockquote>
Thank you!<br>
<br>
<br>
<br>
<pre class="moz-signature" cols="72">--
/Henrik Nord Erlang/OTP</pre>
</body>
</html>