suse standard server 8 crashes after a few days w/kernel panic

[Oskar Teran]

hello,
I just installed suse standard server 8 on several proliant 5500's. One of them (the email server) freezes up after a few days. It responds with pings, but you can't ssh into it or use the suse web admin page. Telneting to the most common ports (25, 143, 80) gets no response.
I looked at the logs and I saw this:

Feb 2 10:32:37 mail kernel: Uhhuh. NMI received. Dazed and confused, but trying to continue
Feb 2 10:32:37 mail kernel: You probably have a hardware problem with your RAM chips
Feb 2 10:32:37 mail kernel: CPU 0: Machine Check Exception: 0000000000000007
Feb 2 10:32:37 mail kernel: Bank 0: f60c600002100800 at 000000003a748980
Feb 2 10:32:37 mail kernel: Kernel panic: CPU context corrupt

Feb 1 20:55:50 mail ctl_deliver[9873]: DBERROR db4: 4 lockers
Feb 1 23:46:17 mail kernel: CPU 0: Machine Check Exception: 0000000000000004
Feb 1 23:46:17 mail kernel: Bank 0: f60f600002000800 at 000000002a58a880
Feb 1 23:46:17 mail kernel: Kernel panic: CPU context corrupt

What do you think this is? Bad RAM sticks?
This box has 400MHZ x 4 CPU's, 1GB RAM, 3 scsi disks setup w/RAID 5. I did all the patches and firmware updates from compaq and suse.

Thanks,

Oskar

[Steven E. Protter]

Its probably RAM.

I got the same message when I stuck a bad RAM chip into a Red Hat Server. The Kernel is the same.

SEP

[Jerome Henry]

Yup,

Try to switch ram sticks from another machine, I had this on a machine when a chip on a stick was burned.

J

[Ragu_1]

Yes, one of the RAM sticks has gone bad. But, how ? Check your power supply units very closely before you replace your bad RAM.