[Ocfs-users] Cannot mount after upgrade OCFS

[Ivan Wong]

Just an update to the good people in the list who has help/responded.
This solution from Sunil actually work for us. We are able to fix the
problem and get the mount point remounted. 


Thanks / regards,
 
Ivan Wong
Database Administrator
 
e2Open Inc. (www.e2open.com) 
Suite 34.03, Level 34, 
Menara Citibank,
156, Jalan Ampang,
50450 Kuala Lumpur, Malaysia
DID: +603 2776 6374 
Tel: +603 2776 6300 
Fax: +603 2712 9112


-----Original Message-----
From: Sunil Mushran [mailto:Sunil.Mushran at oracle.com] 
Sent: Friday, July 28, 2006 1:20 AM
To: Ivan Wong
Subject: Re: [Ocfs-users] Cannot mount after upgrade OCFS


Strange, only the first 8 bytes are off. Minor should be 2 and major 1.

Use the block editor to "fix" the volume.
http://oss.oracle.com/~smushran/.blked/

Familiarize yourself in the readonly mode.
# ./blked.ocfs -o 0 /dev/sdX

Then when you are ready, do:
# ./blked.ocfs -w -o 0 /dev/sdX
and update the block. Only change the major/minor versions.

Ivan Wong wrote:
> Hi experts,
>  
> We did an ocfs upgrade from:
>  
> ocfs-2.4.9-e-smp-1.0.9-9
> ocfs-support-1.0.9-9
> ocfs-tools-1.0.9-9
> to
>  
> ocfs-2.4.9-e-smp-1.0.13-1
> ocfs-support-1.0.10-1
> ocfs-tools-1.0.10-1
> with below enviroment now:
>  
> [oracle at x335-a52 oracle:a52stg04]$ uname -a
> Linux x335-a52 2.4.9-e.25smp #1 SMP Fri Jun 6 18:11:40 EDT 2003 i686
> unknown
> [oracle at x335-a52 oracle:a52stg04]$ rpm -qa | grep ocfs
> ocfs-2.4.9-e-smp-1.0.13-1
> ocfs-support-1.0.10-1
> ocfs-tools-1.0.10-1
> After upgrade, we find one of the mountpoint corrupted:
>  
> [root at x335-a55 root]# mount /ora02
> mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
>        or too many mounted file systems
>
>  
> **********************************************************************
> **********
> [root at x335-a52 root]# fsck.ocfs /dev/sdb1
> fsck.ocfs 1.0.10-PROD1 Fri Mar  5 14:35:41 PST 2004 (build 
> b5602eb387c7409e9f814faf1d363b5b)
> Checking Volume Header...
> WARNING: nonzero bytes after the disk header structure          
> <<<<<<< corruption
> ERROR: structure failed verification, fsck.c, 384
> ocfs_vol_disk_hdr
> =================================
> minor_version: <INVALID VALUE> 0
> major_version: <INVALID VALUE> 2
> signature: OracleCFS
> mount_point: /ora02
> serial_num: 0
> device_size: 26843529216
> start_off: 0
> bitmap_off: 56320
> publ_off: 23552
> vote_off: 39936
> root_bitmap_off: 0
> data_start_off: 1368064
> root_bitmap_size: 0
> root_off: 2416640
> root_size: 0
> cluster_size: 131072
> num_nodes: 32
> num_clusters: 204785
> dir_node_size: 0
> file_node_size: 0
> internal_off: 1368064
> node_cfg_off: 4096
> node_cfg_size: 17408
> new_cfg_off: 21504
> prot_bits: -rwxr-xr-x
> uid: 5550 (oracle)
> gid: 5550 (dba)
> excl_mount: OCFS_INVALID_NODE_NUM
> ERROR: Volume header bad. Exiting, fsck.c, 669              <<<<<<<< 
> corruption
> /dev/sdb1: 2 errors, 0 objects, 0/204785 blocks
>
************************************************************************
********
>  
> We still have 8 more production servers to upgrade to and would
> definitely like to avoid/solve this problem. Please advice.
>  
>  
>  
>
> Thanks / regards,
>
>  
>
> Ivan Wong
>
> Database Administrator
>
>  
>
> e2Open Inc. (www.e2open.com <http://www.e2open.com/>)
>
> Suite 34.03, Level 34, Menara Citibank
>
> 156, Jalan Ampang,
>
> 50450 Kuala Lumpur, Malaysia
>
> DID: +603 2776 6374 Tel: +603 2776 6300 Fax: +603 2712 9112
>
>  
> ----------------------------------------------------------------------
> --
>
> _______________________________________________
> Ocfs-users mailing list
> Ocfs-users at oss.oracle.com 
> http://oss.oracle.com/mailman/listinfo/ocfs-users
>