BL20p G3 repeated power issues

[rossb2]

I have had repeated power issues on a BL20P G3. Initially it was reporting insufficient power from the blade enclosure(just this blade) in the IML logs, and shutting itself down. The status light was stuck on red. Other blades in the same slot were okay.

Under warranty, HP engineers did the following
-system board replaced, insufficient power error still present in IML, although the status light did go green once booted. The server was still shutting itself down intermittently.

-enclosure upgraded to managemenet module firmware 2.40 & blades upgraded to latest ILO firmware 1.92, issue still there.
-Another engineer replaced system board, power converter module & DC filter. The system board failed to boot on the first attempt(no bios output).
-Another system board was brought in the following day. Server worked okay & this cured the insufficent power message in the IML logs. Unfortunately, the following day, the server has had a total power failure. There is now no power input for this blade on any enclosure.

I could get an HP engineer to replace another system board/power components. It could be that HP keep giving me faulty components, but this many power failures is very odd. Anyone else experienced anything like this? Could it be an enclosure issue?

[Sameer_Nirmal]

Yes, we have had 2 similar incidence not too long ago.

The first incidence was very nasty as it would take down the whole enclosure.
The problem was with blade 1 in that case. HP CE replaced the sleeve, signal backplane and updated the enclosure firmware to 2.40 but the power problem was persisting. At last the blade was replaced but it wouldn't power on automatically. But the blade is working normal besides others in the enclosure.

In the second incidence for the same power issue, HPCE replaced the system board and voltage converter but the blade # 3 ( different enclosure) failed after few hours. Then they replaced the sleeve but the blade failed on the next day. They replaced the system board again and since then it's working normal. HP insisted to update the f/w to 2.40 on this enclosure. But that wasn't the cause of the problem.

Knowing that you have blade iLOs and the enclosure at latest f/w level, it concerns me that the problem still might occur.
There is something terribly wrong with the p-class products which under certain conditions goes south. It's hard to pin-point the cause and sometimes the fix doesn't make any sense.


Did you notice the iLO version of this failed blade? I guess there has to be some compatibility between the f/w level of iLO and the enclosure.

Maybe in your case, replacing the system board would work.

[rossb2]

Thanks for yor response. I have had a new sytem board. The blade has power, but will not boot to windows, displaying a "hardware malfunction error".

If I move the blade to an older firmware 2.30 enclosure, it works okay. I tried updating ILO/BIOS again(since hardware replaced), but the problem is still there. Could be an enclosure issue.

I have started a new thread on the hardware malfunction issue, & I will see what i can get from HP support on this.

[Mike Bollman_1]

Ross have you had any luck resolving this issue?

we have the exact same problem (BL25p) and so far have been unable to resolve. We have tried different slots, different enclosures, different racks even.

Same thing, same error.

We got it to boot one time initially, but then it shut itself down after about 30minutes.

Thanks,
Mike B.

[rossb2]

I dont know what is causing the "hardware malfunction error". It only happens on the blades original enclosure slot. At least one other enclosure slot is okay.

[akap475a]

The server (bl20p G3) was working fine until we had to cut the power off and we put out the server from it's position and now it doesn't boot! It has only a red light.

Please HELP!

[Mike Bollman_1]

Ultimately we had to replace the motherboard and CPU's to get the system back online. Which makes absolutely zero sense given that it was working one minute prior.

All is fine now.

[rossb2]

The fault is now fixed.

Issues were
-original system board fault
-faulty replacement system boards

The last fault "hardware malfunction error" when booting windows was not actually a hardware fault, but a software fault. It looks like the original build had problems with newer system board hardware.

I had to rebuild windows using the latest smart start CD. Interestingly, a proliant support pack install prior to that did not fix, implying the proliant support pack was missing at least one driver.

[rossb2]

.