Mailing Lists: Apple Mailing Lists
Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RAID-specific "-9998" errors (enableRAID and addToRAID)

Has anyone else experienced Apple RAID errors, specifically one with the code "-9998"?

I have earlier today but have since been unable to repeat these errors and that's the frustrating part (if I could repeat the steps to create these errors, then I could document what to avoid)!

It all started yesterday when I wanted to (before production) do a dry run of a RAID destruction and restoration process (important to do dry runs before putting a system into production). So, to start things off, I took a perfectly fine RAID mirror (both members had Online status and the mirror itself had Online status according to the results of "diskutil checkRAID"). I then split one of the members off of the RAID and checked the status after the split (still no problem - had Online status for both the RAID and the remaining included mirror), and then I used the diskutil destroyRAID verb to destroy the RAID. Simple as pie, no problems were reported after the destroyRAID verb was executed.

I then erased both of the disc volumes (which were former members of the now deceased RAID volume) with the appropriate file system type which in this case was HFS Case-sensitive Journaled a.k.a. HFSX Journaled. No problem erasing these volumes. I then chose one volume (in the example below the volume associated with the device assigned to "disk1s9") to restore a (compressed, checksummed) disc image to. Restoration was perfect as best I could tell (no complaints from asr), just like this:

# asr -source current-snapshot.dmg -target /dev/disk1s9 -erase -noprompt
Validating target...done
Validating source...done
Erasing target device /dev/disk1s9...done
Retrieving scan information...done
Validating sizes...done
Restoring  ....10....20....30....40....50....60....70....80....90....100
Verifying  ....10....20....30....40....50....60....70....80....90....100

Next, I then attempted to create a new mirror from the just-restored disc volume. I have done this before and the enableRAID verb use does not appear to be rocket science in terms of use and difficulty to understand, so here's what happened:

# diskutil enableRAID mirror disk1s9

Changing filesystem size on disk 'disk1s9'...
Attempting to change filesystem size from 169261555712 to 169261531136 bytes
Filesystem grow failed, 28
Disk Management could not shrink the filesystem to fit the new RAID headers
Error enabling disk to RAID Invalid request (-9998)

I had never experienced such errors before from using enableRAID so I then tried to enableRAID from the other available and opposing disc volume (that was formerly in the deceased RAID mirror), in this case disk2s9. Similarly I restored disk2s9 in the exact same manner as I had to disk1s9 using asr as aforementioned (no problems, the asr restoration succeeded), and then I used enableRAID on disk2s9 to create a new RAID mirror and had no problems doing so:

# diskutil enableRAID mirror disk2s9

Changing filesystem size on disk 'disk2s9'...
Attempting to change filesystem size from 169261555712 to 169261531136 bytes
The filesystem may need to be modified to make this partition bootable
Found new RAID Master
Changing filesystem size on disk 'disk4'...
The disk has been converted into a RAID

Presto! The magic works and disk2s9 has been converted to a RAID mirror (a subsequent diskutil checkRAID verifies that the new RAID mirror (disk4) and the sole member node (disk2s9) have an Online status)! So then it came time to add to this new mirror the other guy, disk1s9 using the addToRAID verb, and Apple's RAID software sure had a bone to pick with disk1s9:

# diskutil addToRAID member disk1s9 disk4
The target disk appears to be too small to add to this RAID set.

There was an error adding the disk to the RAID (-9998)

Rather perplexing, I started to think that the physical disk that contains the device disk1s9 might have a problem with it (although I don't have the diagnostic tools yet to discover if there are physical problems with a disk but I'm investigating such tools like Tech Tool).

Since that time, I just ended up re-creating all of the partitions on disk1 and started from scratch and have since had zero problem creating new RAID volumes using enableRAID with volumes residing on both of my physical discs (and have had no problems in using the destroyRAID and the addToRAID verbs).

I always allow for the high probability that human error (read: me) is to blame (either I didn't understand something fully or I made some other type of error). However, on this score, I wish I could have repeated my error so that I could document the steps to my -9998 errors (and either figure out what I did wrong or possibly report to Apple). Has anyone else run into a wall like this before, with the mystical RAID -9998 error?

Further meta information: My environment for the above context is an Xserve G4 running OS X Server 10.4.9 with a PCI SATA card inside, with two physical external SATA drives (the internal drives are the same both in terms of size, manufacturer, model number, etc.) connected to this machine. This environment has been quite stable for me and has given me hardly any problems.

Thank you for any comparisons / suggestions.

Sergio

 _______________________________________________
Do not post admin requests to the list. They will be ignored.
Macos-x-server mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/macos-x-server/email@hidden

This email sent to email@hidden


Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.