System rebooting itself

[Lacrosse]

About 6 months ago our rp7400 (4x750MHz procs) started rebooting at random times and the core dump always points to processor 3. Per HP support instructions we have replaced the processor same prob then reloaded HPUX 11i with the latest sept 2003 11i release with patch bundle and reloaded Oracle.
HP cannot seem to figure it out they keep thinking software, I am thinking hardware...since we have replaced the processor and it keeps pointing to the same processor my guess would be backplane, any feedback on this would me most appreciated!!

[Patrick Wallek]

Escalate, escalate, escalate!!!!! It's time to get the serious back line support guys and possibly the lab guys involved in your call. Tell the Response center to escalate the call. If you have PSS or higher support, contact you ASE and get them involved as well, that's what they are there for.

It's definitely time to get some heavy duty HP guys involved.

[Lacrosse]

they are already involved but still attcking from a software perspective

[Mark Grant]

We had a similar problem with a V class some time ago. It turned out to be memory. I'm not sure this is relevant to you at all but it might be :)

[Bryan D. Quinn]

Hello,

I have had a similar situation with one of my N's. On a couple of occassions it has rebooted for no apparent reason. I did notice that every once in a while I would get Low DC voltage messages in my GSP. Well, last week it did it again, but this time it did not come back up. I called HP and they had me read off my GSP messages and determined that there was a bad processor power converter. My feeling is that it had been flaky for a while and that is what was giving us random reboots from time to time.

I would definately mention this to HP and keep an eye on your GSP message that complain about Low DC Voltage or High DC Voltage.

-Bryan

[Lacrosse]

Problem resolved the core dump was pointing out Proc#3 in the N4000 I guess Proc#3 resides in slot/position 7 when looking in the N4000 (I guess the HP SE should know this) not slot 3 so HP was replacing the wrong processor
so if this ever happens to you make sure they are replacing the right processor

[Jeff Schussele]

Unless the panic msgs, tombstones & dumps *clearly* point to SW-induced panics, then it will turn out to be HW. Could be backplane, main board, power supply, power monitor board or something.
Like Patrick says escalate this puppy & don't let the pressure off until it's truly solved.

Rgds,
Jeff

[Saurav_1]

Hi,

There is a command called "mca" which comes with the latest support tools. Pls go thru the man page & U can analise the "ts" files in /var/tombstones. The output will show where is the problem. This will work only if U have latest support tools. and Pls attach the latest ts files to ur next reply.

Saurav

[ketan_5]

Even after replacing processor still the same error message is coming then probability with backplane can not be ruled out. Escalate the issue at upper level with both HP and Oracle.

Anyway frankly speaking I do not see any problem with oracle. There must be something with hardware only. It can be with v-regulator/processor/backplane anything.

In our case we observed hanging problem in our K-Class server and oracle made responsible but was wrong and found problem with memory. Involvement of higher level HP technocrates is essential now...