Snap 4200 VOL0 won't mount

[blue68f100]

disk 3, s:0, o:1, n:3 rd:3 us:1 dev:hdg4 kernel Sep 16 03:10:28
disk 2, s:0, o:1, n:2 rd:2 us:1 dev:hde4 kernel Sep 16 03:10:28
disk 1, s:0, o:1, n:1 rd:1 us:1 dev:hdc4 kernel Sep 16 03:10:28
disk 0, s:0, o:1, n:0 rd:0 us:1 dev:hda4 kernel Sep 16 03:10:28

hdg2 is drive 4 by your earlier post hdg4.

Does it not give an indicator led on the front?

[cianwill]

Ok, excellent...well I had cleared out the log and rebooted. Should have figured that one out though, wonder why in one place it is hdg2 and hdg4 in another? Anyway guess it's time to order larger drives -woohoo!

[cianwill]

BTW, I have hooked a KB and Monitor to the 4200. I can watch it go through the boot process, but right after stage 2 it pops up a message stating that it cannot be administered by KVM and video will now be disabled, please connect through the web interface. SO, how the heck are you to do those commands that need interaction locally???

[blue68f100]

Log in to the the web pages and see if there is a option for the monitor & KB. It may be the option when you into maint mode.

Sorry we can't be much help. These have just started coming through here. And I do not have one to run the mill to test on.

[cianwill]

No go on the KB and video, the original documentation that came with it states that the video is disabled. I guess the only way to use them is with some other OS loaded like FreeNAS or Server 2003.

[cianwill]

Oh and the failed drive, I shut down the server and removed the drive to get the s/n to see about the warranty. When I put it back in it was seen as a spare and let me add it back to the array. Still can't get the Volume to load though...
Unbeleavabe, the warranty ran out 7/27/06 - the drives have a mfg date of 7/23/05
that blows!

[Hallis]

Yep, That's why i buy those drives with the 5 year warranties. Too bad no such warranties come with the majority of OEM drives in these servers.

[blue68f100]

Most all oem is 1 yr. aslong os the mfg so they can start sseling parts, Was it A WD or Maxtor drive.... the cheapes of all drives.

I have the 5yr Seagates to, I figure in my 2200 runing 24/7 in about 3 yrs tha y will start failing.The the replacement is only required to cover the ballance of the warraty time.
I have actually have better luck with Hitachi drives lately, even though they only have a 3 yr warranty.

[cianwill]

They came in the server, didn't have a choice. You are right, too bad about the cheapies.They are Maxtor, my 4000 has Seagates with 5yr warranties.

[cianwill]

XFS mounting filesystem lvm(58,0) kernel Sep 22 00:05:36
XFS: extended permissions ON kernel Sep 22 00:05:36
XFS: log mount failed kernel Sep 22 00:05:29
XFS: log mount/recovery failed kernel Sep 22 00:05:29
XFS: failed to locate log tail

Replaced HD and still not able to mount Vol0, above is the latest from the logfile.

[blue68f100]

I'm afraid that something failed to be written to the HD when the power outage hit.

Do you have the means to connect and FD or CD drive to the unit.
If so you could run a disk utility on the drives. I know SpinRite supports the XFS file structure.

Do you know what OS version you have, and do you have it where you can reload it?

I'm running out of options.

Search and/or post on Gentoo forum, or google search that error. I know you were not the first to get it.

Or you may need to break down and call adaptec $$$$

[blue68f100]

This is a different error than before. Which tells me the log maybe corrupted on the original hd. Putting the old drive back and try repairing the log.

Doing a search on google there are multiple hits for "failed to locate log tail".
http://oss.sgi.com/archives/xfs/2002-11/msg00386.html said to try "xfs_repair -L" to repair the log. THIS FALLS UNDER SERIOUS XFS DAMAGE.

One user reported reloading the kernel corrected the problem. Which re-installing the gsu may corect the probem. You do have the CD or the GSU file Dont you. If not contacting adaptec may be your only option.

*** Since I have no experience in the Guardian OS, I would advise contacting Adaptec and pay for support if you must have the data, and you have no backup or snapshots.

But please keep up posted if you find solution. So we may learn and pass it on to others.

[cianwill]

Ok,

Here is what I got when I ran these two commands...any insight?

xfs_repair -L /dev/hda4

watchdog set to 255
Phase 1 - find and verify superblock...
bad primary superblock - bad magic number !!!
attempting to find secondary superblock...
found candidate secondary superblock...
error reading superblock 19 -- seek to offset 81604378624 failed
unable to verify superblock, continuing...
found candidate secondary superblock...
error reading superblock 19 -- seek to offset 81604378624 failed
unable to verify superblock, continuing...




xfs_repair -L /dev/md0

Phase 1 - find and verify superblock... bad primary superblock - bad magic number !!!
attempting to find secondary superblock...
.................................................f ound candidate secondary superblock...
unable to verify superblock, continuing...


I am going to look on gentoo

Thanks for the link....

[jontz]

Hmmm...."bad primary superblock - bad magic number"

That's not good. When XFS mounts a file system, it looks at the Superblock first and then the object location table. If it can't find and/or verify the superblock, something in the very beginning of your file system has been overwritten or corrupted. I don't know that there is much you can do to fix this on your own, as the traditional method for fixing this sort of problem is a format and I know you want to keep the data on the drives.

I'll do a little more research and let you know what I come up with...

[Hallis]

both I and some of my friends have fairly good linux skills. I might be able to repair it if in fact it will mount. The Data pertition might be lost im afraid. It would be helpful if you could make an image of the drive.

[cianwill]

I was able to hook a usb cdrom to it and boot off a knoppix cd. I was able to run
e2fsck but I still get bad superblock errors and the volume still won't mount. Still tinkering...

[blue68f100]

Looks like I need to do some more research on recovery from superblock failure. So far I'm out of my league on this one. But these are major problems that the Guardian OS addressed with SnapShot Image files. The negative part is its eats up some of raid storage space.

Cainwill:

What did the Adaptec's help support suggest when you talked to them?

[cianwill]

Adaptec want's 200$ for 1 year Bronze package - yeah right! I wonder if I were to upgrade it to a newer version (software) if that would somehow fix the superblock error - probably not. Just theorizing - and that's IF I can lay my hands on a newer OS.

[Phoenix32]

You're not kidding... Please let us know if you do...

[blue68f100]

I have read that some recovery utilities can rebuild the superblock. And I would bet most all of the recovery services have them.

Since you can boot from a external source. May do some research on recovery software. There was another one that said it rebuild with the set in place if capable of spinning up.

Rule #1 :
Make a clone of the drives. So you have a set to work on and preserve the originals.

[re3dyb0y]

Originally Posted by blue68f100

I have read that some recovery utilities can rebuild the superblock. And I would bet most all of the recovery services have them.

Since you can boot from a external source. May do some research on recovery software. There was another one that said it rebuild with the set in place if capable of spinning up.

Rule #1 :
Make a clone of the drives. So you have a set to work on and preserve the originals.

David, give me a shout if you need a hand with it

[blue68f100]

I guess when ever i get setup and start testing. I'm going to have to corrupt the superblock to see if any of the utilites I have will repair it.

re3dyb0y, If I were to shout you would never hear me with all the water between us. But thanks for the offer.

[re3dyb0y]

*Sighs*

You know what i mean :P

[blue68f100]

of course. Your skills are always welcome.