RP8400 2 npars

[Juan M Leon]

Hi during the attempt to recover root password the npar0 was RS'ed or TC'ed to boot in single user mode. However the server never displayed anything after the attempted recycle. Then we checked the VFP and found the following.
Cell Status
0 OS reboot Shutdown entered
1 heartbeat

We executed PS for cell 0 from the MP prompt. The output was the following

Power status: not viable
boot is blocked
processor compatibility: fault
RIO cable status : undefined

We attempted to RS cell 0 again and the output reported RtnCodeNoPoweredCellSlost in Partition

Attempted to Enable "ON" the Cell and Partition and we were unable, the status remained "OFF".
We TC'ed the npar1 and halted
The last step was to reset the MP and power cycle the entire cabinet. After the power was recycled both npars came back ok. The status was heartbeat for both npars.

Has anybody experienced the same problem before?

[John Payne_2]

It sounds like you had a power supply problem. In, PS, when you look at the cabinet (option t), do you currently have redundant power?

Total Power Available 6000 VA
Total Power Needed 4008 VA
Power Redundancy : redundant
Power Viability : viable

Maybe you currently do not have redundant power, and you have a flaky power supply. We certainly have had some.

the PS command is the right place to be to diagnose your problem, you just have to look at the rest of the output to see what is failed next time. (Since you hard booted, you probably just masked the problem until another time...)

Hope it helps

John

[generic_1]

No, to seen before.
I would update your firmware and see if that seems to fix it.
I would comb the logs for any odd errors in syslog, tombstones, and in the GSP.
Open a call with support, see if they have seen it before.
It woulnt hurt to have a CE look inside to make sure things are seated well and that there are no module mismatches. Like in the N class hardware, just an example, there were certain modules that had to be installed for a certain number of processers/speed I believe. Sometimes they have patch/required modules to be installed with certain hardware. It would not hurt to check. There are also multiple revisions of harware designs often too. This may not be true in your case but it cant hurt to check.

[Juan M Leon]

Thanks for your input John, do you believe is a hardware problem? we have power redundancy. as per the cabinet I dont remenber the exact Power available/needed numbers but there was information on that output.
Assuming that there is problem with my power sources is there any way I can diagnostic the problem without waiting for the next time to happen. Also remember that the rp8400 has 2 npars and only one of them was hang after the RS. The second npar did not experienced any problems.

Thanks
Juan

[Juan M Leon]

John, another information that I didnt entered in my original message is the following:
VRM's cell 0
Present
Enabled
Power Good

The output was empty no information.

Thank you

[Juan M Leon]

anybody have any more ideas.

thank you

Juan

[Juan M Leon]

I found the following Thread where it refers that the MP reflects unknown state for the nparts and to get updated status the server was powercycled:
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=708127
My situation is similar, only one npar remains in shutdown reboot status after attempt to reboot. Now a second rp8420 behave the same way 2 days ago.

Anyone has any suggestions.

Thank you