how to prevent filesystem check

Andreas Dilger adilger at clusterfs.com
Wed Nov 29 05:20:26 UTC 2006


On Nov 28, 2006  12:33 -0800, Wolber, Richard C wrote:
> Running the following command on your slave server should do the trick:
> 
> echo "AUTOFSCK_DEF_CHECK=\"no\"" >> /etc/sysconfig/autofsck

This is incorrect.  As soon as the ext3 code mounts the filesystem
it will do journal recovery and potentially corrupt the filesystem.
Then, the read-only copy will become out-of-date in the cache of
that client and it will get bogus data back, eventually deciding
that the filesystem is corrupt (whether it is or not).

You should just mount the filesystem on the client via NFS, that's
what it's SUPPOSED to do.

This is a good reason for the multi-mount protection feature that I
proposed previously.  It would mark the filesystem as in-use on one 
node and the filesystem itself would refuse to mount on the second
node.  Unfortunately, this idea met resistance from some of the
other ext3 developers from merging it upstream.

Cheers, Andreas
--
Andreas Dilger
Principal Software Engineer
Cluster File Systems, Inc.




More information about the Ext3-users mailing list