dmesg time stamp in the future?

370 views
Skip to first unread message

Alan Orth

unread,
May 16, 2014, 7:22:19 AM5/16/14
to nairo...@googlegroups.com
So I have this server running Tomcat and some other Java stuff, and the
kernel's OOM killer just killed a java process. The time stamp is
unreadable:

[38949106.375446] Out of memory: Kill process 23179 (java) score 781 or
sacrifice child

Apparently that's an offset from the time the machine booted. But
`dmesg -T` shows:

[Tue Jul 21 22:57:49 2015] Out of memory: Kill process 23179 (java)
score 781 or sacrifice child

How can it be in the future... the server's time is correct (NTP).
Anyone ever seen this? As there are are several Java-based applications
running here, it kinda makes it hard to troubleshoot exactly which
`java` was killed.

The server is an Ubuntu 12.04 server, in case you were wondering.

--
Alan Orth
alan...@gmail.com
http://alaninkenya.org
http://mjanja.co.ke
"I have always wished for my computer to be as easy to use as my telephone; my wish has come true because I can no longer figure out how to use my telephone." -Bjarne Stroustrup, inventor of C++
GPG public key ID: 0x8cb0d0acb5cd81ec209c6cdfbd1a0e09c2f836c0


signature.asc

Dennis Mungai

unread,
May 16, 2014, 7:46:55 AM5/16/14
to Nairobi Linux Users Group
Alan,

I've encountered that before with a mis-behaving Glassfish server in Ubuntu 12.04.
Back then, we were playing around with Dataverse DVN, and one of its' install scripts kept re-spawning new Glassfish server instances in an endless loop.

Never paid too much attention to it, though.
--
Please avoid sending me Word or PowerPoint attachments.

See http://www.gnu.org/philosophy/no-word-attachments.html

Dennis Mungai

unread,
May 16, 2014, 2:20:32 PM5/16/14
to Nairobi Linux Users Group

alias dmsg_human = dmesg|perl -ne ‘BEGIN{$a= time()- qx!cat /proc/uptime!};s/\[(\d+)\.\d+\]/localtime($1 + $a)/e; print $_;’

dmesg is not human friendly when it comes to time stamps. For some distros, dmesg -T works.

Alan Orth

unread,
May 16, 2014, 3:34:36 PM5/16/14
to nairo...@googlegroups.com
Since we're off topic, this also works in modern distros:

dmesg --human -T

Alan

On 05/16/2014 09:20 PM, Dennis Mungai wrote:
> alias dmsg_human = dmesg|perl -ne ‘BEGIN{$a= time()- qx!cat
> /proc/uptime!};s/\[(\d+)\.\d+\]/localtime($1 + $a)/e; print $_;’
>
> dmesg is not human friendly when it comes to time stamps. For some
> distros, dmesg -T works.
>
> On 16 May 2014 14:46, "Dennis Mungai" <dmn...@gmail.com
> <mailto:dmn...@gmail.com>> wrote:
>
> Alan,
>
> I've encountered that before with a mis-behaving Glassfish server in
> Ubuntu 12.04.
> Back then, we were playing around with Dataverse DVN, and one of
> its' install scripts kept re-spawning new Glassfish server instances
> in an endless loop.
>
> Never paid too much attention to it, though.
>
>
> On 16 May 2014 14:22, Alan Orth <alan...@gmail.com
> <mailto:alan...@gmail.com>> wrote:
>
> So I have this server running Tomcat and some other Java stuff,
> and the
> kernel's OOM killer just killed a java process. The time stamp is
> unreadable:
>
> [38949106.375446] Out of memory: Kill process 23179 (java) score
> 781 or
> sacrifice child
>
> Apparently that's an offset from the time the machine booted. But
> `dmesg -T` shows:
>
> [Tue Jul 21 22:57:49 2015] Out of memory: Kill process 23179 (java)
> score 781 or sacrifice child
>
> How can it be in the future... the server's time is correct (NTP).
> Anyone ever seen this? As there are are several Java-based
> applications
> running here, it kinda makes it hard to troubleshoot exactly which
> `java` was killed.
>
> The server is an Ubuntu 12.04 server, in case you were wondering.
>
> --
> Alan Orth
> alan...@gmail.com <mailto:alan...@gmail.com>
> http://alaninkenya.org
> http://mjanja.co.ke
> "I have always wished for my computer to be as easy to use as my
> telephone; my wish has come true because I can no longer figure
> out how to use my telephone." -Bjarne Stroustrup, inventor of C++
> GPG public key ID: 0x8cb0d0acb5cd81ec209c6cdfbd1a0e09c2f836c0
>
>
>
>
>
> --
> Please avoid sending me Word or PowerPoint attachments.
>
> See http://www.gnu.org/philosophy/no-word-attachments.html
>
> --
> You received this message because you are subscribed to the Google
> Groups "Nairobi GNU/Linux User Group" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to nairobi-gnu...@googlegroups.com
> <mailto:nairobi-gnu...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.
signature.asc
Reply all
Reply to author
Forward
0 new messages