Re: vinum, fsck generates errors when Plex 1 is up

2004-08-29 Thread Peter A. Giessel
> * What problems are you having?

fsck reports dozens and dozens of errors and the filesystem as corrupt
whenever I start Plex 1.

> * Which version of FreeBSD are you running?

5.2.1-RELEASE

> * Have you made any changes to the system sources, including Vinum?

No, I recompiled the kernel a couple of times, most recently to remove
USB so that the following stops (and after removing USB it did stop):

Aug 18 09:07:32 raid51 kernel: usb0: 1 scheduling overruns
Aug 18 09:07:55 raid51 last message repeated 465 times

> * Supply the output of the vinum list command. If you can't start
> Vinum, supply the on-disk configuration, as described below. If you
> can't start Vinum, then (and only then) send a copy of the
> configuration file.

It is currently as follows, after
# vinum stop array.p1
12 drives:
D two   State: up   /dev/ad18s1hA: 0/190732 MB (0%)
D one   State: up   /dev/ad16s1hA: 0/190732 MB (0%)
D four  State: up   /dev/ad15s1hA: 47/190779 MB (0%)
D five  State: up   /dev/ad14s1hA: 47/190843 MB (0%)
D three State: up   /dev/ad12s1hA: 47/190843 MB (0%)
D twelveState: up   /dev/ad11s1hA: 47/190779 MB (0%)
D ten   State: up   /dev/ad10s1hA: 47/190779 MB (0%)
D elevenState: up   /dev/ad8s1h A: 47/190779 MB (0%)
D eight State: up   /dev/ad7s1h A: 47/190779 MB (0%)
D nine  State: up   /dev/ad6s1h A: 47/190843 MB (0%)
D seven State: up   /dev/ad5s1h A: 0/190732 MB (0%)
D six   State: up   /dev/ad4s1h A: 0/190732 MB (0%)

1 volumes:
V array State: up   Plexes:   2 Size:931 GB

2 plexes:
P array.p0   R5 State: up   Subdisks: 6 Size:931 GB
P array.p1   R5 State: faulty   Subdisks: 6 Size:931 GB

12 subdisks:
S array.p0.s0   State: up   D: one  Size:186 GB
S array.p0.s1   State: up   D: two  Size:186 GB
S array.p0.s2   State: up   D: threeSize:186 GB
S array.p0.s3   State: up   D: four Size:186 GB
S array.p0.s4   State: up   D: five Size:186 GB
S array.p0.s5   State: up   D: eleven   Size:186 GB
S array.p1.s0   State: obsolete D: six  Size:186 GB
S array.p1.s1   State: obsolete D: sevenSize:186 GB
S array.p1.s2   State: down D: eightSize:186 GB
S array.p1.s3   State: down D: nine Size:186 GB
S array.p1.s4   State: down D: ten  Size:186 GB
S array.p1.s5   State: down D: twelve   Size:186 GB

The following was the "vinum l" from after "start array.p1" finished:
12 drives:
D two   State: up   /dev/ad18s1hA: 0/190732 MB (0%)
D one   State: up   /dev/ad16s1hA: 0/190732 MB (0%)
D four  State: up   /dev/ad15s1hA: 47/190779 MB (0%)
D five  State: up   /dev/ad14s1hA: 47/190843 MB (0%)
D three State: up   /dev/ad12s1hA: 47/190843 MB (0%)
D twelveState: up   /dev/ad11s1hA: 47/190779 MB (0%)
D ten   State: up   /dev/ad10s1hA: 47/190779 MB (0%)
D elevenState: up   /dev/ad8s1h A: 47/190779 MB (0%)
D eight State: up   /dev/ad7s1h A: 47/190779 MB (0%)
D nine  State: up   /dev/ad6s1h A: 47/190843 MB (0%)
D seven State: up   /dev/ad5s1h A: 0/190732 MB (0%)
D six   State: up   /dev/ad4s1h A: 0/190732 MB (0%)

1 volumes:
V array State: up   Plexes:   2 Size:931 GB

2 plexes:
P array.p0   R5 State: up   Subdisks: 6 Size:931 GB
P array.p1   R5 State: up   Subdisks: 6 Size:931 GB

12 subdisks:
S array.p0.s0   State: up   D: one  Size:186 GB
S array.p0.s1   State: up   D: two  Size:186 GB
S array.p0.s2   State: up   D: threeSize:186 GB
S array.p0.s3   State: up   D: four Size:186 GB
S array.p0.s4   State: up   D: five Size:186 GB
S array.p0.s5   State: up   D: eleven   Size:186 GB
S array.p1.s0   State: up   D: six  Size:186 GB
S array.p1.s1   State: up   D: sevenSize:186 GB
S array.p1.s2   State: up   D: eightSize:186 GB
S array.p1.s3   State: up   D: nine Size:186 GB
S array.p1.s4   State: up   D: ten  Size:   

Re: vinum, fsck generates errors when Plex 1 is up

2004-08-22 Thread Greg 'groggy' Lehey
On Sunday, 22 August 2004 at 11:35:15 -0800, Peter A. Giessel wrote:
> I have two plexes in an array (FreeBSD 5.2.1-RELEASE, which I needed so
> that I could run Samba3 with ACL).
>
> The following is roughly how my array is (was) set up.
> volume array
>   plex org raid5  (p0)
> drive one  (p0.s0)
> drive two  (p0.s1)
> drive three(p0.s2)
> drive four (p0.s3)
> drive five (p0.s4)
> drive eleven   (p0.s5)
>   plex org raid5  (p1)
> drive six  (p1.s0)
> drive seven(p1.s1)
> drive eight(p1.s2)
> drive nine (p1.s3)
> drive ten  (p1.s4)
> drive twelve   (p1.s5)

That's impossible.  You can't put plexes on drives; they've got to be
on subdisks, which you don't mention.  This is why I ask
(http://www.vinumvm.org/vinum/how-to-debug.html) for output from
vinum(8), which I see you added below:

> 2 plexes:
> P array.p0   R5 State: up   Subdisks: 6 Size:931 GB
> P array.p1   R5 State: faulty   Subdisks: 6 Size:931 GB
>
> 12 subdisks:
> S array.p0.s0   State: up   D: one  Size:186 GB
> S array.p0.s1   State: up   D: two  Size:186 GB
> ...
> S array.p1.s0   State: staleD: six  Size:186 GB
> S array.p1.s1   State: obsolete D: sevenSize:186 GB
> ...

It would be nice to know when this was done; I assume it was after the
successful fsck.

Still, to continue...

> All was well for a while, then drive twelve failed, so i went to
> replace it, created a configfile:
> drive twelve device /dev/ad11s1h
>
> # vinum create configfile
> # vinum start array.p1.s5
>
> when p1.s5 finished reviving, I got all kinds of fsck errors such
> as "INCORRECT BLOCK COUNT", "EXCESSIVE BAD BLOCKS", etc.

It would be interesting to know whether this was with the file system
mounted and active or not.  There have been some race conditions which
might cause this problem.  If the file system was mounted, could you
please try again with it unmounted?  If it wasn't, the information in
the URL above would helpful.

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
Note: I discard all HTML mail unseen.
Finger [EMAIL PROTECTED] for PGP public key.
See complete headers for address and phone numbers.


pgpjxtz5phHvQ.pgp
Description: PGP signature


vinum, fsck generates errors when Plex 1 is up

2004-08-22 Thread Peter A. Giessel
I have two plexes in an array (FreeBSD 5.2.1-RELEASE, which I needed so
that I could run Samba3 with ACL).

The following is roughly how my array is (was) set up.
volume array
  plex org raid5  (p0)
drive one  (p0.s0)
drive two  (p0.s1)
drive three(p0.s2)
drive four (p0.s3)
drive five (p0.s4)
drive eleven   (p0.s5)
  plex org raid5  (p1)
drive six  (p1.s0)
drive seven(p1.s1)
drive eight(p1.s2)
drive nine (p1.s3)
drive ten  (p1.s4)
drive twelve   (p1.s5)

All was well for a while, then drive twelve failed, so i went to
replace it, created a configfile:
drive twelve device /dev/ad11s1h

# vinum create configfile
# vinum start array.p1.s5

when p1.s5 finished reviving, I got all kinds of fsck errors such
as "INCORRECT BLOCK COUNT", "EXCESSIVE BAD BLOCKS", etc.

Fortunately, I had run:
# fsck -n /dev/vinum/array
so it didn't mess everything up.  I then tried:

# vinum stop array.p1
then I ran:
# fsck -n /dev/vinum/array

It was clean, no errors.  I tried restarting array.p1:
#vinum start array.p1

However, whenever I start that plex (and its done reviving), I get tons
of fsck errors.  Did I do something I shouldn't have done?  Did I try
to do something with vinum that it can't do?

Is there any way to fix it?  Any suggestions? TIA.

Here is my "vinum list" output:
12 drives:
D two   State: up   /dev/ad18s1hA: 0/190732 MB (0%)
D one   State: up   /dev/ad16s1hA: 0/190732 MB (0%)
D four  State: up   /dev/ad15s1hA: 47/190779 MB (0%)
D five  State: up   /dev/ad14s1hA: 47/190843 MB (0%)
D three State: up   /dev/ad12s1hA: 47/190843 MB (0%)
D twelveState: up   /dev/ad11s1hA: 47/190779 MB (0%)
D ten   State: up   /dev/ad10s1hA: 47/190843 MB (0%)
D elevenState: up   /dev/ad8s1h A: 47/190779 MB (0%)
D eight State: up   /dev/ad7s1h A: 47/190779 MB (0%)
D nine  State: up   /dev/ad6s1h A: 47/190843 MB (0%)
D seven State: up   /dev/ad5s1h A: 0/190732 MB (0%)
D six   State: up   /dev/ad4s1h A: 0/190732 MB (0%)

1 volumes:
V array State: up   Plexes:   2 Size:931 GB

2 plexes:
P array.p0   R5 State: up   Subdisks: 6 Size:931 GB
P array.p1   R5 State: faulty   Subdisks: 6 Size:931 GB

12 subdisks:
S array.p0.s0   State: up   D: one  Size:186 GB
S array.p0.s1   State: up   D: two  Size:186 GB
S array.p0.s2   State: up   D: threeSize:186 GB
S array.p0.s3   State: up   D: four Size:186 GB
S array.p0.s4   State: up   D: five Size:186 GB
S array.p0.s5   State: up   D: eleven   Size:186 GB
S array.p1.s0   State: staleD: six  Size:186 GB
S array.p1.s1   State: obsolete D: sevenSize:186 GB
S array.p1.s2   State: obsolete D: eightSize:186 GB
S array.p1.s3   State: obsolete D: nine Size:186 GB
S array.p1.s4   State: obsolete D: ten  Size:186 GB
S array.p1.s5   State: obsolete D: twelve   Size:186 GB
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"