HP 9000 735 with 10.20

[Jeff Tolly]

The 735 starts file checking on its own and have not figured out what logs to look at to determine what's causing the behavior.

Also do you have some commands that can be used to diagnose the hard drive or at least give some indications of what's not working correctly on the system. Sorry, not a Unix guru

This server will be replaced in a month but need it to run until then.

Thank you

[Sanjay_6]

Hi Jeff,

start by looking at /var/adm/syslog/syslog.log

You may need the help of this doc.

http://docs.hp.com/hpux/onlinedocs/5187-2216/5187-2216.html

Hope this helps.

Regds

[Kent Ostby]

Jeff ...

For each disk on your system run:

dd if=/dev/rdsk/ of=/dev/null bs=64k

If you find any disks that report I/O error or error number 5 then this is your problem disk.

Likewise if you look at "dmesg" output, and you find scsi errors with a bdev --> , you can compare the last 5 digits of the hex value to an ll /dev/dsk and locate the problem disk.

Best regards,

Kent M. Ostby

[Bill Hassell]

When you say "file checking" do you mean the fsck program? This program won't start by itself. Perhaps someone has put this program into cron? (crontab -l to see what is running automatically) Or do you see fsck running after the system stops responding and you see a bunch of text scrolling up the screen on a black background? If so, the system has crashed and you're seeing a reboot due to the crash. In that case, you need to look at the ts99 file in /var/tombstones to see if an HPMC occurred. You can also use q4 to analyze the crash dump (if a dump was saved). If it's hardware, you'll just have to live with it for the month. Otherwise you probably need patches: https://payment.ecommerce.hp.com/cgi-bin/swdepot_parser.cgi/cgi/try.pl?productNumber=B3782XSW1&date=

[Jeff Tolly]

Could not find a Tombstone dir