Proliant DL380 G3 Smart Array 5i "waiting for rebuild"

[Jason Lehrhoff]

I saw my third drive showing as failed so I reseated it. This happened once in a blue moon. Now it shows as rebuilding (flashing green drive icon) but array ends up showing in the ACU (version 8.x)as "awaiting rebuild".
I have rebooted the server, reseated the drive and even tried putting another drive in there (bigger 72GB Ultra320 15k) just temporarily until I can get a replacement 18.2GB drive, but now the problem is back on this drive as well so maybe it is my controller? Can I skip that slot and put it in slot 3 ?

Report Attached.

[Gary Antonio Benavides]

Run the Array Diagnostic Utility Report start/all programs/HP System tool. Make sure there is not a second HDD in predictive failure or with read errors. Sometimes it doesn't let a new hard drive to rebuild the information. On the Array Configuration Utility, click on the storage controller, then on common tasks and check the rebuild priority.

[Jason Lehrhoff]

I set the rebuild and expand priorities to high. I think what is happening is one of the other drives has a read error that is not allowing the rebuild of the added drive.
If this is the case, I think I am kind of screwed as far as getting this array up and running without wiping it out and starting again.
If it has 3 disks and one is in a rebuild "queue" and another has errors, then I can't even pull another drive to replace it with a good one. I attached the ADU report

[Jason Lehrhoff]

I basically used Ghost to clone the server to an external USB drive (if anyone can avoid doing this and do it over ethernet do it for USB 1.1 takes 2 hours to do this!)

I pulled the bad drive, then populated all of the open slots with 18.2GB SCSI Ultra3 15k drives. The RAID rebuilt but the drive in slot 0 has no activity light flashing (the green arrow light that usually flashes same time on all drives.). This was suspect to me, esp. since this was the same slot reporting errors previously. So I pulled the drive the next day and it rebuilt but now I get the "background parity initialization is queued..." message. This message has been up for 2 days now whereas it usually goes thru this process in a few hours. I even set the rebuild priority to high with no luck.

I am going to pull the drive again and replace it with one of the spares I ordered. I hope this fixes the issue. If not, is it possible it is the slot that has the problem and not the drive (since 2 drives so far are acting sketchy in slot 0)

Attaching the current ADU report (report2.zip)

[Jason Lehrhoff]

I'm also hoping it is ok to pull this drive while that "background parity" message is displayed...if anyone knows the impact of this let me know.

[MT19]

Yes, it could be the slot, so I would definitely consider replacing the SCSI backplane in that server. I've noticed quite a number of SCSI backplane issues with DL380 G3s at our site and have had to have them replaced.