Lost disklabel - How to write new label - More information regard ing to my previous question

From: Padiyath Sreekumaran (Kumar.Padiyath@psi.ch)
Date: Mon May 02 2005 - 09:50:33 EDT


      Hello,
      I have more information regarding my given problem. I had 3 defect
disks
      which I replaced.From 3 domain in the HSZ40 attached raid disks I
could
      mount 2 domains without errors. The third one which I need(where the
disks
      are replaced) gives me the following error during the reboot:(This
domain
      cannot be mounted)

      May 2 08:54:46 psw224 vmunix: ADVFS: using 2322 buffers containing
18.14 megabytes of memory May 2 08:55:31 psw224 vmunix: AdvFS I/O error:
May 2 08:55:31 psw224 vmunix: Volume: /dev/rz28c
May 2 08:55:31 psw224 vmunix: Tag: 0xfffffff3.0000
May 2 08:55:31 psw224 vmunix: Page: 49
May 2 08:55:32 psw224 vmunix: Block: 880
May 2 08:55:32 psw224 vmunix: Block count: 256
May 2 08:55:32 psw224 vmunix: Type of operation: Read
May 2 08:55:32 psw224 vmunix: Error: 5
May 2 08:55:32 psw224 vmunix: AdvFS I/O error:
May 2 08:55:32 psw224 vmunix: Volume: /dev/rz28c
May 2 08:55:32 psw224 vmunix: Tag: 0xfffffff3.0000
May 2 08:55:32 psw224 vmunix: Page: 33
May 2 08:55:32 psw224 vmunix: Block: 624
May 2 08:55:32 psw224 vmunix: Block count: 256
May 2 08:55:32 psw224 vmunix: Type of operation: Read
May 2 08:55:32 psw224 vmunix: Error: 5
May 2 08:55:37 psw224 vmunix: AdvFS I/O error:
May 2 08:55:37 psw224 vmunix: Volume: /dev/rz28c
May 2 08:55:37 psw224 vmunix: Tag: 0xfffffff3.0000
May 2 08:55:37 psw224 vmunix: Page: 1085
May 2 08:55:37 psw224 vmunix: Block: 17456
May 2 08:55:37 psw224 vmunix: Block count: 256
May 2 08:55:37 psw224 vmunix: Type of operation: Read
May 2 08:55:37 psw224 vmunix: Error: 5
May 2 08:55:37 psw224 vmunix: AdvFS I/O error:
May 2 08:55:37 psw224 vmunix: Volume: /dev/rz28c
May 2 08:55:37 psw224 vmunix: Tag: 0xfffffff3.0000
May 2 08:55:37 psw224 vmunix: Page: 1069
May 2 08:55:37 psw224 vmunix: Block: 17200
May 2 08:55:38 psw224 vmunix: Block count: 256
May 2 08:55:38 psw224 vmunix: Type of operation: Read
May 2 08:55:38 psw224 vmunix: Error: 5
May 2 08:55:38 psw224 vmunix:
May 2 08:55:38 psw224 vmunix: load_x_cache: bad status from bs_refpg of sbm
May 2 08:55:38 psw224 vmunix:
May 2 08:55:38 psw224 vmunix: AdvFS Domain Panic; Domain data_dom Id
0x386a27a5.00051a10 May 2 08:55:38 psw224 vmunix: An AdvFS domain panic has
occurred due to either a metadata write error or an inter nal inconsistency.
This domain is being rendered inaccessible. May 2 08:55:38 psw224 vmunix:
Please refer to guidelines in AdvFS Guide to File System Administration
regarding w hat steps to take to recover this domain.

      Which utility can I try? I am trying salvage -D directoryname -V
/dev/rz28h which gives the
      following error:

salvage: I/O error
salvage: read() failed
salvage: I/O error
salvage: read() failed
salvage: I/O error
salvage: read() failed
salvage: I/O error
salvage: read() failed
salvage: I/O error
salvage: read() failed
salvage: I/O error
salvage: read() failed
salvage: I/O error

      I can read the label. There are 11 disks in the domain. Is there any
utility with which
      I can get part of the information from the disks?

      Sorry for the trouble and thanks in advance.

      Regards,
      Kumar

>
> Hello,
> I have a system with the following data:
>
> AlphaServer 3800
> OS V4.0D
> Connected via HAZ40 to a diskarray.
>
> The system(Server only) was not in operation the last
> 6 months. Bur the diskarray was always switched on.
> The battery(HSZ) is okay. I can see the disk array from boot prompt
> as well as after booting. There are 2 members in ase. (Failover
> configuration). The second server is not started. The problem is
> I have lost the disklabel of the disks in the raid configuration.
> Part of the
> uerf command output:
>
> tza1 firmware version: DEC TZA
> _SPO/A.14
> scsi3 at tza1
> rz25 at scsi3 target 1 lun 0
> (LID=4)
>
> _() (Wide16)
> rz26 at scsi3 target 2 lun 0
> (LID=12)
> _() (Wide16)
> rz27 at scsi3 target 3 lun 0
> (LID)
> _() (Wide16)
> rz28 at scsi3 target 4 lun 0
> (LID=28)
> _() (Wide16)
> tcds1 at tc0 slot 3
>
>
> The disklabel is corrupted. Iam using advanced file system. When I
execute
> "disklabel -r rz25", I get the following error:
>
> disklabel: /dev/rrz25a: I/O error
>
> I donot have no backup since we have moved to AFS File system. All
> the the
> user directories are moved to AFS. Now one user who does not have
> an AFS account
> likes to have one of his program.The question is How can I relabel
> the disk
> without destroying the datas on the disk?
>
> Any help is very much appreciated.
------------------------------------------------------------------
Padiyath Sreekumar | Tel: +41.56.310.3643
Paul Scherrer Institut | email: kumar.padiyath@psi.ch
AIT | Office: WHGA/U132
WHGA/U132 | Fax: +41.56.310.3649
CH-5232 Villigen PSI |
Switzerland |
-----------------------------------------------------------------
------------------------------------------------------------------
Padiyath Sreekumar | Tel: +41.56.310.3643
Paul Scherrer Institut | email: kumar.padiyath@psi.ch
AIT | Office: WHGA/U132
WHGA/U132 | Fax: +41.56.310.3649
CH-5232 Villigen PSI |
Switzerland |
-----------------------------------------------------------------



This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:50:18 EDT