vmunix log events

[gany59]

Hi,

the server which runs on 11.11, is generating millions of vmunix log events by this the host has been producing a great deal of messages that impact OVO syslog collector. 7->8 million log events per day.

Could any body tell me how to fix this problem or how to stop the logs

Thanks in advance...
Points assured!!!!

[Patrick Wallek]

>>Could any body tell me how to fix this problem or how to stop the logs

Well, the easiest way is to fix whatever problem is causing the vmunix events to be generated in the first place.

Are they all the same event? Are there different events?

A **small** sample of what is being logged will be helpful in troubleshooting.

[gany59]

Hello Patrick,

Thanks for ur response..
the below is the some lines of the logs output

syslog-20091114.gz:7966549
syslog-20091115.gz:7918412
syslog-20091116.gz:8048748
syslog-20091117.gz:7950438
syslog-20091118.gz:7914826
syslog-20091119.gz:8063633
syslog-20091120.gz:7932193
syslog-20091121.gz:7902417

[Dennis Handly]

>the below is the some lines of the logs output
syslog-20091114.gz:7966549

This is the content of syslog.log? These look like files and sizes?

What occurs before and after that pattern of entries?

[Bill Hassell]

These messages do not look like vmunix messages. kernel messages will usually have the word vmunix. To get rid of the messages, I would restart syslogd with the -v option. That will now add the priority and facility code to every message. I suspect that this is some script or application and hopefully it uses one of the local facilities in which case, that facility can be filtered out of the log.

You still need to track down the bad code. Start by looking for scripts that create or manipulate gzip (.gz) files. You may have to capture extended ps output and then search the processes to see if gzip shows up:

UNIX95=1 ps -fH >> /tmp/ps.out

Run this 10 or 20 times (perhaps every few seconds) and then vi the file looking for gzip or gunzip. The -H option will show the process tree. You are looking for the parent of the gzip processes.

[Bill Hassell · Accepted solution]

Make that:

UNIX95=1 ps -efH >> /tmp/ps.out

The -e will capture everything.

[gany59]

Hi actuall the vm unix log is below:

Dec 1 07:58:18 alicaz vmunix: function == 0x7AA050, arg == 0x40F4A800, ticks == 0xD26FAD40, flags == 0x0
Dec 1 07:58:18 alicaz vmunix: function == 0x7AA050, arg == 0x40F4A800, ticks == 0xD26FAD3F, flags == 0x0
Dec 1 07:58:18 alicaz vmunix: function == 0x7AA050, arg == 0x40F4A800, ticks == 0xD26FAD3E, flags == 0x0
Dec 1 07:58:18 alicaz vmunix: function == 0x7AA050, arg == 0x40F4A800, ticks == 0xD26FAD3D, flags == 0x0
Dec 1 07:58:18 alicaz vmunix: function == 0x7AA050, arg == 0x40F4A800, ticks == 0xD26FAD3C, flags == 0x0

i am getting more logs like this.. but i am not able to determine what was the error..

Thanks in advance !!!!!!!!

[Johnson Punniyalingam]

Check the below thread,

Refering to "firmware" upgrade,

looks like its bug,

http://forums13.itrc.hp.com/service/forums/questionanswer.do?admit=109447627+1259655530650+28353475&threadId=1386609

[Dennis Handly]

>I am getting more logs like this.. but I am not able to determine what was the error.

As Johnson said, it's a bug. Your system has been up too long and it's getting tired. ;-)