Xserve RAID error codes

Question

Level 1

38 points

Xserve RAID error codes

Is there a list of error codes for the Xserve RAID, specifically disk errors, which look something like:

Timestamp:	28/01/16 01:55:43 PM
Priority:	Warning
Controller:	Upper Controller
Type:	112
Event ID:	543
Event:	Disk 5 Reported An Error. COMMAND:0x42 ERROR:0x40 STATUS:0x51 LBA:0x3FA80700
Description:	The drive reported an ATA error. This is a failure in the communication from the RAID Controller to the drive.

I am guessing that these are S.M.A.R.T. errors, but I cannot find a list of codes that has any of these. These are Seagate drives, if it makes any difference.

TIA.

Xserve RAID A1009 with Seagate 750GB HDDs.

Safari-OTHER

Posted on Jan 29, 2016 5:37 AM

Reply

Answer 1

Jan 29, 2016 8:48 AM in response to Clive Bruton

Here's the official overview:

RAID Admin: Understanding Logical Block Address (LBA) messages in the event log - Apple Support

Based on the format and contents, those are probably ATA/ATAPI commands and errors. Look in the T13.Org ATA/ATAPI documents for details on that stuff, and see if anything matches. The other common choice here is the T10.org SCSI documentation.

Occasional disk errors and disk replacements are expected. That's why you're running RAID, after all. You can try to scan the drive and the array, and/or to swap it, but the Xserve RAID FC SAN storage array itself is old and unsupported and probably due for a more general replacement.

Reply

Answer 2

Clive Bruton Author

Level 1

38 points

Jan 29, 2016 4:23 PM in response to MrHoffman

Thanks, that link specifies one of the error messages (bad block), but not the other. I'm guessing that it cannot be that serious (right now) because it's not triggering an amber light warning on the drive module. Sure, I expect to get errors, but it's nice to be informed and understand what those errors mean. If I find a reference I'll link it back here. As to upgrading… right now it seems that you can buy at least 10-15 Xserve RAIDs for the price you'd pay for something approaching parity (just in terms of capacity, let alone redundancy) and around 100 of them for the sort of price required for a "modern" RAID… and there seem to be plenty of parts around, so no problem keeping them going (if you don't need raw speed).

Reply