https://bugs.dragonflybsd.org/https://bugs.dragonflybsd.org/favicon.ico?16293952082005-12-15T00:58:20ZDragonFlyBSD bugtrackerDragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=452005-12-15T00:58:20Zdillon
<ul></ul><p>:hi folks,<br />:<br />:sometimes sysctl or fstat just stop and end up as a zombie process<br />:<br />:% ps<br />: PID TT STAT TIME COMMAND<br />: 81 #C1 SL 0:00.04 _su -m (csh)<br />:40884 #C1 IWL 0:00.00 sysctl<br />:41510 #C1 DWL 0:00.00 sh -c /usr/bin/fstat > /dev/null 2>&1<br />:41513 #C1 ZL 0:00.00 /usr/bin/fstat<br />:53544 #C1 IWL 0:00.00 sysctl<br />:53549 #C1 IWL 0:00.00 sysctl<br />:53647 #C1 DL 0:00.01 sh -c /usr/bin/fstat > /dev/null 2>&1<br />:53648 #C1 ZL 0:00.00 /usr/bin/fstat<br />:53651 #C1 DL 0:00.02 sh -c /usr/bin/fstat > /dev/null 2>&1<br />:53654 #C1 ZL 0:00.00 /usr/bin/fstat<br />:57387 #C1 R1L+ 0:00.01 ps<br />:<br />:I can't kill them, even with kill -9<br />:<br />:so my question is, ignore them or is smth broken?</p>
<pre><code>Well, fstat can seg-fault trying to access kernel memory. The<br /> question is why aren't these zombies being reaped by their parent.</code></pre>
<pre><code>If you generate a kernel core dump with the system in this state<br /> and upload it to leaf I'll try to track it down.</code></pre>
<pre><code>-Matt<br /> Matthew Dillon <br /> &lt;<a class="email" href="mailto:dillon@backplane.com">dillon@backplane.com</a>&gt;</code></pre> DragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=482005-12-15T18:48:30Zskrueger
<ul></ul><p>Matt suggested turning off the mpsafe sysctl's (they're turned off by<br />default) and this worked, no more zombie processes, no more strange<br />netstat -m output, no systat -ifstat errors</p> DragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=512005-12-16T05:59:20Zdillon
<ul></ul><p>:..<br />:> Well, fstat can seg-fault trying to access kernel memory. The<br />:> question is why aren't these zombies being reaped by their parent.<br />:> <br />:> If you generate a kernel core dump with the system in this state<br />:> and upload it to leaf I'll try to track it down.<br />:Matt suggested turning off the mpsafe sysctl's (they're turned off by<br />:default) and this worked, no more zombie processes, no more strange<br />:netstat -m output, no systat -ifstat errors</p>
<pre><code>Right. The mpsafe sysctls are experimental. I put them in there so<br /> we could start identifying areas that need work, and we've identified<br /> probably half a dozen issues so far. They aren't turned on by default<br /> and so will not effect the release. The areas in question, along with<br /> other goodies, are going to be worked on after the release.</code></pre>
<pre><code>In the case of fstat the issue is simply the process p_lock counter<br /> being accessed without appropriate MP protection.</code></pre>
<pre><code>-Matt<br /> Matthew Dillon <br /> &lt;<a class="email" href="mailto:dillon@backplane.com">dillon@backplane.com</a>&gt;</code></pre> DragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=13642006-10-02T20:16:17Zcorecode
<ul></ul><p>is this fixed?</p> DragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=14132006-10-05T00:09:00Zdillon
<ul></ul><p>:Simon 'corecode' Schubert <<a class="email" href="mailto:corecode@fs.ei.tum.de">corecode@fs.ei.tum.de</a>> added the comment:<br />:<br />:is this fixed?</p>
<pre><code>I think this one is fixed. Hopefully all the lockf related bugs have<br /> been fixed now.</code></pre>
<pre><code>-Matt<br /> Matthew Dillon <br /> &lt;<a class="email" href="mailto:dillon@backplane.com">dillon@backplane.com</a>&gt;</code></pre> DragonFlyBSD - Bug #14: zombie processes waiting for a lock, smth to worry about?https://bugs.dragonflybsd.org/issues/14?journal_id=14672006-10-17T04:52:02Zskrueger
<ul></ul><p>sry for replying so late</p>
<p>yes this bug is fixed and the issue can be closed</p>
<p>thx</p>