Issues mirroring drives with Vinum in FreeBSD 4.8-RELEASE

[aarong]

On Saturday, October 4, 2003, at 10:18  PM, Greg 'groggy' Lehey wrote:

> On Saturday,  4 October 2003 at 10:36:15 -0700, aarong wrote:
>> On Saturday, October 4, 2003, at 01:30  AM, Greg 'groggy' Lehey wrote:
>> fstab: /etc/fstab:9: Inappropriate file type or format
>
> It's probably worth fixing this.

I hosed that particular installation and reinstalled; it complains no 
more.

>> # fsck -n /dev/vinum/var
>> ...
>> ** Phase 5 - Check Cyl groups
>> FREE BLK COUNT(S) WRONG IN SUPERBLK
>> SALVAGE? no
>
> This can be normal.
>
>> SUMMARY INFORMATION BAD
>> SALVAGE? no
>
> So can this.

Then neither of the above would cause a drop into single user mode, I 
gather?

>> ALLOCATED FRAG 1277319 MARKED FREE
>> BLK(S) MISSING IN BIT MAPS
>> SALVAGE? no
>>
>> ALLOCATED FRAG 1368488 MARKED FREE
>> ...same message until 1368493...
>
> But these suggest something worse.  The problem here is that it found
> the superblock, so it's likely that your geometry is correct: you
> wouldn't have got this far if you hadn't.
>
>> fsck'ing the rest of the volumes produces more of the same; one 
>> notable
>> difference is root has a reference count error in addition to missing
>> bit maps, bad summary information, and incorrect number of block 
>> counts
>> in the superblock. The usr volume seems to be fine according to fsck,
>> however. vinum list correctly notes both drives are up, four volumes
>> have two plexes, the 8 plexes are up and sized, and the subdisks are
>> up. fsck -n /dev/vinum/var before mirroring produced no complaints,
>> otherwise I'd conclude that I just mirrored a corrupt plex but that
>> doesn't seem to be the case.
>
> It looks something like that.  In single user mode, do an fsck on each
> of the component plexes.  My guess is that (at least) one plex of each
> volume will be bad.

I was not aware you could fsck individual plexes; I'll definitely try 
that on this fresh installation. Thank you for your time and insight, 
Greg; I'll be sure to send an update when I figure out what went wrong 
the first time.

-aarong

> Greg
> --
> When replying to this message, please copy the original recipients.
> If you don't, I may ignore the reply or reply to the original 
> recipients.
> For more information, see http://www.lemis.com/questions.html
> See complete headers for address and phone numbers.
> <mime-attachment>