DL360 Original, disk problems

[Oleg Koroz]

Port 2 SCSI ID 1 that report predictive failure ?
from ADU both drives looks good, at least for me

had you ever replaced power supply ?
or any other parts as ROC (Array module or system board)

[ernie cline]

I inherited the system, not sure if the power supply has been replaced, but since its a G0 DL360, it probably has, since I think they all failed at some point :).

the SMH reports the disks as 'box s, port 101' so i can't tell which disk is actually the one that thinks its in predictive failure, but as soon as i get that alert, 5 minutes later I get the 'OK' alert so ...

[Oleg Koroz]

Errors that you have from ADU looks hardware related
If you have one more server with no problem, try to swap SCSI backplane, after System board; you still can try Power Supply and small power filter at the back
Or reset SCSI backplane for now

I don't think ROC will cause that

PLS reply back after you find problem

It's possibly that Drive from ID 1 has some Spin up problem

[Gus Kwong]

Hi Ernie,

If the drive is really about the die, you can try and create extra load on the disks. I ususally use the SQL stress test tool from microsoft (just search for SQLIOStress.exe) and it can really hammer your disks. It has managed to kill some of my predictive-failure disks.

Gus

ps: if this forum has been useful please assign points.

[Basil Vizgin]

Also you can try to check with Smart Array Drive Diagnosis utility from HP Insight Diagnostics Online Editionhttp://h18004.www1.hp.com/products/servers/management/hpid/

[ernie cline]

HPDiags report success. Also I don't think the disks are actually dying, I think there is just some kind of software problem, maybe the driver interface to the array controller ... i've had this problem before but updating the firmware and BIOS always corrects it. Is there some way to reset the error count on these disks?

I can't run the SQL server test, this is Linux :).

HPDiags says the array controller is fine ...