Server Crash

[unixdaddy]

One of my servers keeps crashing. In the OLDsyslog.log it gives the information below. Can anyone tell me what it means?

Oct 7 15:26:13 arvdev vmunix: cdb: 00 00 00 00 00 00
Oct 7 15:26:13 arvdev vmunix: status: (02) Check Condition
Oct 7 15:26:13 arvdev vmunix: sense data: 70 00 06 00 00 00 00 0e 00 00 00 00
29 00 00 00
Oct 7 15:26:13 arvdev vmunix: 00 00 00 00 00 00
Oct 7 15:26:13 arvdev vmunix: sense key: (06) Unit Attention
Oct 7 15:26:13 arvdev vmunix: additional sense code: (29)
Oct 7 15:26:13 arvdev vmunix: additional sense code qualifier: (00)
Oct 7 15:26:27 arvdev /usr/sbin/envd[991]: terminated by signal 15
Oct 7 10:26:37 arvdev inetd[1522]: Going down on signal 15

[Stefan Farrelly]

I have no idea what the driver arvdev is, some sort of 3rd part product ?? anyway, the errors you see look like SCSI hardware connection errors pointing to a problem with the device, eventually it causes a crash. Are your patches uptodate ? Keeping your patches uptodate usually stops the crashes, but youve got to investigate the hardware connection problem also.

[Marcin Wicinski]

Hi,
There can be other problems (indicated by inetd going down on signal 15), if you have trusted system. If so, try to install also patches for inetd (PHNE_21835).

Later,

[unixdaddy]

arvdev is actually the name of the server. I think your probably right. I wonder why it occurs because it just reboots and comes back up and everything is fine for a while.

[Bill McNAMARA_1]

are you getting a dump in /var/adm/crash
when the server reboots.

If so you could run
# q4 -p /var/adm/crash/crash.X
q4> run WhatHappened > /tmp/wh.X
q4> quit
# vi /tmp/wh.X
and try to search the patch db for stack trace id's... or send that to your support with an swlist output

Later,
Bill

[unixdaddy]

I'm afraid there is nothing in /var/adm/crash.

[Mark Vollmers]

Darren-

I don't suppose that you could be lucky and have it always go down at the same time? You should be able to get times from /etc/shutdownlog. Also, do you have Online diagnostics installed (stm)? If not, maybe you should put it on. It a free download or it's on the application cd. If there is a hardware issue, it should give some warnings or errors that might help you track it down.

Mark

[Sanjay_6]

Hi Darren,

Can you attach your OLDsyslog.log file. Let us have a look at that. Seems like you are having a disk problem. Which disk (?) is a big question.

Thanks

[E. Wong]

Darren,

It may be some change in your computer room temperature. An envd entry in your log may indicate whatever environmental conditions you have specified in /etc/envd.conf (OVERTEMP) file. Are you in a hot climate location?

E.