Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 13 Jun 2002 07:59:44 -0500
From:      "Jack L. Stone" <jackstone@sage-one.net>
To:        Layder <layder@ukr.net>, freebsd-stable@FreeBSD.ORG
Subject:   Re: Zombie processes (fwd)
Message-ID:  <3.0.5.32.20020613075944.01219690@mail.sage-one.net>
In-Reply-To: <20020613131834.S4010-100000@cyanide.pro>

next in thread | previous in thread | raw e-mail | index | archive | help
At 01:24 PM 6.13.2002 +0300, Layder wrote:
>
>
>---------- Forwarded message ----------
>Date: Wed, 12 Jun 2002 15:29:30 -0500
>From: Jack L. Stone <jackstone@sage-one.net>
>To: Layder <layder@ukr.net>, freebsd-stable@FreeBSD.ORG
>Subject: Re: Zombie processes
>
>At 11:00 PM 6.12.2002 +0300, Layder wrote:
>>Processes in my box never gone. They become zombies and their
>>quantity increased with time and than locks the system talking that
>>cannot fork process. What can I do?
>>My system is Celeron 500 with 64 MB RAM
>>
>Here was the advice I received a couple of weeks ago:
>  $ ps alx | grep Z
>   UID   PID  PPID CPU PRI NI   VSZ  RSS WCHAN  STAT  TT       TIME COMMAND
>   1004 82075  6901 295  96  0     0    0 -      Z+    p5    0:00.00  (sh)
>
>See the Z under STAT?  That show that the process is a zombie.
>
> and what does it mean...????
>
>It means that the process has died, but hasn't been buried yet.  The
>trouble with zombies is that you can't kill them, because they're
>already dead :-) First the parent process (6901 in this example) must
>wait for it and collect its status.
>
>You have two choices here:
>
>1.  Ignore it, it's not holding on to too many resources.
>2.  Kill the parent.  That may or may not be a good idea.  For
>    example, in this case,
>
>    $ ps lp6901
>      UID   PID  PPID CPU PRI NI   VSZ  RSS WCHAN  STAT  TT       TIME
COMMAND
>     1004  6901  6620 127  20  0   848  102 opause S+    p5    5:42.68
>observe dates
>
>    This is a process I have on my machine which shows the dates
>    around the world every 2 seconds.  Sure, I can kill it and restart
>    it, but is there any point?  In this case, if I go back and look
>    for that zombie, I'll find it's already gone: I've just caught a
>    zombie in the stage before being buried.  In other cases, and
>    especially if the program is causing lots of zombies, you may want
>    to stop it.
>
>Best regards,
>Jack L. Stone,
>Administrator
>
>SageOne Net
>http://www.sage-one.net
>jackstone@sage-one.net
>
>
>-----------------------------------------
>
>That's what ps -jfax | grep Z shows:
>root      62     1    62 c0a5e840    0 Z     ??    0:00.00  (ppp)
>root     177     1   177 c0a89d80    0 Z     ??    0:00.00  (squid)
>root     230     1   226 c0a8df80    0 Z     ??    0:00.00  (gzip)
>root     276     1   272 c0a89080    0 Z     ??    0:00.00  (gzip)
>root     322     1   318 c0a89e00    0 Z     ??    0:00.00  (gzip)
>root     372     1   368 c0a914c0    0 Z     ??    0:00.00  (gzip)
>root     435     1   435 c0a5ee40    0 Z     ??    0:00.00  (ppp)
>root     175     1     7 c0a43f80    0 Z    con-   0:00.00  (squid)
>
>after i gzip something, gzip becomes a zombie and never dies completely
>init is the parent of these processes so i can't kill it.
>
>I'll have a bunch of zombies just within a few hours after system startup.
>Can this be a hardware problem? This happens only on one this machine,
>two others work just fine.
>

Not sure why the zombies are created on that one machine, but, it at least
doesn't appear that any are using up resources, so until that happens, it
may not be critical while you look for the source(s) of the problem....

On my machines, only one zombie keeps recurring and that is from the daemon
apcupsd that runs my APC-UPS battery stuff..... they never go away unless I
kill/restart the parent.

Best regards,
Jack L. Stone,
Administrator

SageOne Net
http://www.sage-one.net
jackstone@sage-one.net

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.5.32.20020613075944.01219690>