fsck performed on GFS/GFS2 file systems during boot which caused files moved to lost+found, failure to mount file systems, failure to boot, or data corruption in a RHEL 5 or 6 Resilient Storage cluster
Issue
- We lost some files as a result of GFS2 filesystem corruption. We see some files under the
lost+found
directory. - There are files in the
lost+found
directory with timestamps that match full cluster restarts. Is restarting the cluster causing files to go missing from the filesystem? - Does
GFS2
orGFS
supportfsck
in/etc/fstab
when the host boots up? - Why does my
GFS
orGFS2
filesystem appear to hang when the host is booting? - Cluster nodes will not boot, getting
fsck_gfs2 : invalid option -- a
error. - A cluster node failed to boot, referencing errors when mounting GFS2 or performing an
fsck
and dropped me to a repair shell - Cluster node attempts to filesystem check GFS2 on boot and fails, then drops to a shell
Environment
- Red Hat Enterprise Linux (RHEL) 5 or 6 with the Resilient Storage Add On
- A Global Filesystem (
GFS
) or Global Filesystem 2 (GFS2
) file systems that is listed in/etc/fstab
with a 6th field greater than 0:
/dev/datavg/datalv /var/data/APPDATA gfs2 defaults 0 0
- A Global Filesystem (
GFS
) or Global Filesystem 2 (GFS2
) file systems that is listed in/etc/fstab
with the mount option_netdev
:
/dev/datavg/datalv /var/data/APPDATA gfs2 _netdev 1 2
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.