OCFS2 mounted.ocfs2 returns 'Unknown: Bad magic number in inode' (Doc ID 431556.1)

Last updated on JUNE 22, 2017

Applies to:

Linux OS - Version Oracle Linux 4.4 and later
Linux x86-64
Linux x86
Linux ItaniumLinux x86-64
Oracle Cluster File System Version 2 (OCFS2) 1.x

Symptoms

When running OCFS2 mounted.ocfs2(8) command to check which cluster nodes have OCFS2 volumes mounted, message 'Unknown: Bad magic number in inode' is returned for one one or more devices. Repartitioning (fdisk(8)) or reformatting (mkfs.ocfs2(8)) the device(s) fail to suppress the message.

[root@anubis ~]# mounted.ocfs2 -f
Device                FS     Nodes
/dev/sdc              ocfs2  Unknown: Bad magic number in inode
/dev/sdc1             ocfs2  arachnid.au.oracle.com, anubis.au.oracle.com
/dev/sdd1             ocfs2  arachnid.au.oracle.com, anubis.au.oracle.com

[root@anubis ~]# mounted.ocfs2 -d
Device                FS     UUID                                  Label
/dev/sdc              ocfs2  c9b56493-efb0-4016-9e77-03df29c5b90a  orahome
/dev/sdc1             ocfs2  2f0bf126-94db-4384-9fd4-9cc03d784ef3  orahome
/dev/sdd1             ocfs2  2823dc87-f60d-4d8f-b281-c3c73eb35265  oradata

Changes

The device(s) for which the message is observed, was originally an unpartitioned disk, upon which an OCFS2 filesystem had once been created. The device(s) were subsequently partitioned (whether with or without an OCFS2 filesystem having been created on one or more of the it's partitions).

In the example above, an OCFS2 filesystem was originally created on unpartitioned device /dev/sdc - it was later partitioned and an OCFS2 filesystem created on partition one, /dev/sdc1.

Cause

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms