<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body><table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>Bug ID</th>
<td><a class="bz_bug_link
bz_status_NEW "
title="NEW - [EFL][GTK] Script process-linux-coredump is not needed, switch back to use the kernel core dumper instead."
href="https://bugs.webkit.org/show_bug.cgi?id=160927">160927</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[EFL][GTK] Script process-linux-coredump is not needed, switch back to use the kernel core dumper instead.
</td>
</tr>
<tr>
<th>Classification</th>
<td>Unclassified
</td>
</tr>
<tr>
<th>Product</th>
<td>WebKit
</td>
</tr>
<tr>
<th>Version</th>
<td>WebKit Nightly Build
</td>
</tr>
<tr>
<th>Hardware</th>
<td>Unspecified
</td>
</tr>
<tr>
<th>OS</th>
<td>Unspecified
</td>
</tr>
<tr>
<th>Status</th>
<td>NEW
</td>
</tr>
<tr>
<th>Severity</th>
<td>Normal
</td>
</tr>
<tr>
<th>Priority</th>
<td>P2
</td>
</tr>
<tr>
<th>Component</th>
<td>Tools / Tests
</td>
</tr>
<tr>
<th>Assignee</th>
<td>webkit-unassigned@lists.webkit.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>clopez@igalia.com
</td>
</tr>
<tr>
<th>CC</th>
<td>bugs-noreply@webkitgtk.org, lforschler@apple.com, pnormand@igalia.com
</td>
</tr>
<tr>
<th>Blocks</th>
<td>128928
</td>
</tr></table>
<p>
<div>
<pre>On r203674 and r203677 a script was added to write to disk the coredumps generated by the WebKit tests.
However, this script is not really needed, because the kernel itself can take care of writing the coredumps to disk directly.
Relying on the kernel core dumper to write the cores to disk has some advantages:
* Simplifies the handling of coredumps.
* Avoids running as root a script.
* The coredump file gets written to disk as the user running the tests, rather than as root.
* Avoids potential race conditions. On the WebKit Debug bot I found that sometimes when GDB was executed the coredump was still not 100% fully written to disk, and therefore GDB failed with an error saying that the core was truncated. Probably this could be fixed by setting the sysctl core_pipe_limit to a non-zero value. But to seems simpler and safer not piping to a program/script if its not really needed. Since I switched the GTK Debug bot to use the kernel coredumper rather than piping to the script process-linux-coredump I have not found this issue again.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>