Absolute nightmare of a DL380 G7

[ap_4]

We purchased all the HP Servers all at once from same wholesaler.

The 5630 processors model were affected.
HP has replaced them with 5680 model.

I am not certain if its a faulty piece of hardware or the way they were put together.

I am not aware of any errata.

[ap_4]

These are our errors in the IML:

CPU 11/30/2010 03:06 11/30/2010 03:06 1 Uncorrectable Machine Check Exception (Board 0, Processor 1, APIC ID 0x00000001, Bank 0x00000004, Status 0xB2000000'00020001, Address 0x00000000'00000000, Misc 0x00000000'00000000)
5 CPU 11/30/2010 03:06 11/30/2010 03:06 1 Uncorrectable Machine Check Exception (Board 0, Processor 1, APIC ID 0x00000000, Bank 0x00000004, Status 0xB2000000'00020001, Address 0x00000000'00000000, Misc 0x00000000'00000000)

[Michael A. McKenney]

So HPs solution was replace all the CPUs. I would be curious to find out the reasoning? Where they mismatched in the server? Different stepping?

[GrandSlam]

Sorry, my client server w/ the same problem should be DL360 G7.

Is that the problem occurs due to old iLo3 or SmartArray P410i firmware w/ bug?

Or the problem about CPU as some netfriends reported?

[Jesse Short]

Has anyone had the problems when using the E5640 Cpu's? Just got this server online with 3 virtual machines on it, sure would ruin my day if this might become a problem.

Cheers!

[Michael A. McKenney]

I would check with HP to see what the Errata problems are with it. If in doubt, switch CPUs. When I think a power supply, RAM or CPU are the issue, I just replace them. I don't wait for problems.

[JUGGANUTZ]

That is Crazy! i have the DL385 G7 with the flash memory as well and i see it on my p410i. So somehow i don't think the raid card locking up is associated with the CPU's being bad. Especially since yours are the intel. The firmware corrected one of my servers but i am still on a up hill battle. I just changed the memory so we shall see what happens. You got to love technology.

[BigLupu]

ProLiant DL380 G7 here hanging approx. one a week with this:

Uncorrectable Machine Check Exception (Board 0, Processor 2, APIC ID 0x00000035, Bank 0x00000004, Status 0xB2000000'00020001, Address 0x00000000'00000000, Misc 0x00000000'00000000)
Uncorrectable Machine Check Exception (Board 0, Processor 2, APIC ID 0x00000034, Bank 0x00000004, Status 0xB2000000'00020001, Address 0x00000000'00000000, Misc 0x00000000'00000000)

We'll update to the latest firmware tonight and see if it helps.

[Michael A. McKenney]

Check to see if both CPUs are the same stepping and which stepping they are. Could be a CPU mismatch causing it. It could be a firmware issue.

[BigLupu]

Thanks. Upgraded firmwares last night like this:

iLO: 1.15 -> 1.16
ROM: 2010.09.30 -> 2010.12.01
P410i: 3.52 -> 3.66

It didn't help at all as it took only six hours to machine check exception (attached) happen again.

I checked the stepping to be same on the both CPU:s. They are:

Model: Intel(R) Xeon(R) CPU L5640 @ 2.27GHz
Stepping: 2

I guess we should try to change motherboard, CPU:s (tried this already once) and memory. It really is a nightmare.