[Ocfs2-users] Group Descriptor # 0 has bad signature

[Rich Amick]

Testing OCFS2 on a 2 node cluster.  
Both nodes are HP DL380G5 with RHEL4U3 x86_64 (2.6.9-34.ELsmp).  
OCFS2 version 1.2.5-1.  
Shared storage is MSA1000.

Formatted a 142GB volume with:
mkfs.ocfs2 -b 4K -C 32K -N 4 -L u01 /dev/sda1 

Mounted the formatted partition on both nodes:
mount -t ocfs2 /dev/sda1 /u01

Create a file on oracle1:
[root at oracle1 bin]# echo test > /u01/test2

The file is visable on both the nodes:
[root at oracle0 bin]# ls -l /u01/test2
-rw-r--r--  1 root root 5 Jun  6 13:22 /u01/test2
[root at oracle1 bin]# ls -l /u01/test2
-rw-r--r--  1 root root 5 Jun  6 13:22 /u01/test2

Created a file on oracle0:
[root at oracle0 bin]# echo test > /u01/test3
echo: write error: Input/output error

OCFS2 marks the file system as read-only on oracle0.
/var/log/messages from oracle0:
Jun  6 11:07:05 oracle0 kernel: OCFS2: ERROR (device sda1):
ocfs2_check_group_descriptor: Group Descriptor # 0 has bad signature
Jun  6 11:07:05 oracle0 kernel: File system is now read-only due to the
potential of on-disk corruption. Please run fsck.ocfs2 once the file system
is unmounted.

Unmount the partition on both nodes and run fsck:
[root at oracle0 bin]# /sbin/fsck.ocfs2 /dev/sda1 

Checking OCFS2 filesystem in /dev/sda1:
  label:              /u01
  uuid:               64 ef 7f 3a b7 03 4f 8e 8b e2 82 aa 5b c2 cc ea 
  number of blocks:   35563880
  bytes per block:    4096
  number of clusters: 4445485
  bytes per cluster:  32768
  max slots:          4

/dev/sda1 is clean.  It will be checked after 20 additional mounts.

Remounting the volume reproduces same scenerio.