PE2950 RAID 5 disk failure

Question

My server is a PE2950 with a PERC 6/i controller. I have 3 300GB drives in a RAID 5 config.

The other day I noticed an LED on one of the drives and dug a little deeper and found that drive failure was predicted so I ordered a new drive. Yesterday I shut the machine down, put the new drive in and configured it as a hot spare. I then ran the "replace" command on the failing drive. That process completed last night. I removed the failed drive and restarted the machine this morning and now i am noticing that the activity LED on the new drive is on solid but the other drives that LED is just blinking. I'm unable to find the meaning of that.

Did I do this correctly? Is there something else I should have done or still need to do?

Thanks,

JG

theflash1932 · Answer

In OMSA, does it show your RAID 5 as Optimal/Ready, or does it show it as Degraded? What version of OMSA? What versions are your PERC driver and firmware at?

john_guerra · Answer

Thank you for replying.   OMSA version is 7.1 PERC driver is 4.05.01.64 PERC firmware is 6.3.1-0003 RAID 5 State shows as 'Ready'.

theflash1932 · Answer

You should be fine then ... not all disks will be active at any given time, but if it NEVER shows any activity on the LED, then it may be the disk compatibility (firmware) ... what model disk did you use to replace the PF disk?

theflash1932 · Answer

You don't necessarily need to purchase a drive FROM Dell ... it just needs to be a Dell-branded/certified drive. For example:

From Dell:

<ADMIN NOTE: Broken link has been removed from this post by Dell>

From Server Supply:
http://www.serversupply.com/MFGR/DELL/HARD%20DRIVES%20W-TRAY/SATA-II/1000GB-7200RPM/341-7526.htm

Exact same drive, just from a supplier (which is usually cheaper than buying directly from Dell).

theflash1932 · Answer

Is the new Seagate a 'Dell' drive?  That model is sold BOTH as a retail/generic drive AND a Dell-certified/OEM drive.  The Dell-branded drive will have a specially-tuned firmware loaded on it so as to 'certify' it for use on the PERC's - allowing it to work/communicate seamlessly with the various OEM settings of the PERC.  While generic drives usually work, they also may experience issues ranging from benign or mild (none at all, unknown responses in the logs, inaccurate LED status) to serious or even critical (drives randomly timing out/falling offline, degraded performance, not being configurable, not even being recognized), leading to data loss/corruption.  So, if this is not a Dell drive, then it may continue to work ok, but the status LED would be something you'd have to live with, and hopefully you wouldn't experience any other issues down the road (if you do, this should be your first suspect).  Or you could get a certified drive.

john_guerra · Answer

The replacement disk was as close to the originals as I could find. The originals are Seagate 15K RPM 300GB Drives and the replacement is the same, just a newer model I would assume. The model number of the originals is ST3300655SS and the new one is ST3300657SS.

In OMSA, the older drives report more info than the new one. The new one does not show mfr date info or part number, but the old ones do. I checked Seagate's site for firmware update for the new drive and there are none available.

john_guerra · Answer

After doing some more looking around on this topic, I kind of came to the same conclusion. I did not buy it from Dell. I bought it from newegg. I didn't think of the issues you raised when i bought it, but they make sense now.

This is not a production box, at least not in the sense that users are accessing it for anything. Due to it's age, it is an admin server that handles things like central A/V administration and backups. I'll keep an eye on it and if it starts acting up, I'll buy a drive from Dell.

Thanks for your input. I really appreciate it.

PowerEdge HDD/SCSI/RAID

PE2950 RAID 5 disk failure

Was this post helpful?