strange Filesystem issue

[joe_91]

i have this error from syslog
Jun 13 13:46:19 crusader vmunix: msgcnt 117 vxfs: mesg 008: vx_direrr: vx_dirscan_2 - /m01 file system inode 15 block 7252350 erro
r 6

/m01 is fine( i am able to read/write to it) but a particular dir under /m01/data/oradata has issues. if i do ls -i on oradata it shows 15. it is a empty directory and i am not able to rm -r it also..this is the error i am getting
# rm -r oradata
rm: cannot stat oradata: No such device or address
# I am getting no such device error when i try to copy to this directory also. Only two days back the dame filesystem /m01 had issues and i fsck'd and fixed it. it was a different directory and now this directory..please help with this..i want to atleast remove this directory and re-create with the same name(so that it has diff inode)..any ideas?

THanks

Joe

[Robert-Jan Goossens · Accepted solution]

Hi Joe,

I think your only option is to umount the filesystem and run a full fsck on it.

# fsck -o full /dev/vgXX/rlvolyy

Regards,
Robert-Jan

[spex]

Joe,

Try this:

find /m01/data -type d -inum 15 -exec rmdir -f {} \;

PCS

[Deoncia Grayson_1]

you can read this link for better explanation of your error:

http://www1.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&admit=-682735245+1150226580639+28353475&docId=200000076281156

WARNING: msgcnt x: vxfs: mesg 008:
vx_direrr: caller_id - mount_point file system inode inumber block blkno error
errno
A directory operation failed in an unexpected manner. caller_id identifies the
routine that generated the message. The mount point, inode, and block number
identify the failing directory. If the inode is an immediate directory, the
directory entries are stored in the inode, so no block number is reported. If
the error is ENOENT or ENOTDIR, an inconsistency was detected in the directory
block. This inconsistency could be a bad free count, a corrupted hash chain, or
any similar directory structure error. If the error is EIO or ENXIO, an I/O
failure occurred while reading or writing the disk block.

The VX_FULLFSCK flag is set in the super-block so that fsck will do a full
structural check the next time it is run.

.Action

Check the console log for I/O errors. If the problem was caused by a disk
failure, replace the disk before the file system is mounted for write access.
Unmount the file system and use fsck to run a full structural check.

[Michael Steele_2]

To remove dirs or files by inode is what you seek.

# ls -i * -or- ls -i /rm01

will return ####, ie, 1316

# find / -inum ### -exec rm -i {} \:

[Steven E. Protter]

Shalom Joe,

In summary, its either get fsck to work or build a new filesystem and restore the data from backup.

SEP