On Tue, Oct 29, 2013 at 05:54:46PM -0700, Mathew Snyder wrote:
> As for the scripting, we don't push out any commands such as shutdown. The
> scripts simply report which servers have mounts in an unexpected RO state.

Probably you know this - on RHEL5/6 you can restore RW with

  # mount <mountpoint> -oremount,rw

which is safe and tested (by me) on intact ext2/ext3 filesystems.
To assess if remounting RW would make sense, you *might* be able 
to fsck non-shared mounted RO filesystems.

Our practice is to separate filesystems by function (/tmp,
/var/, /boot, /u00 (Oracle data), etc., to narrow the
focus to "important" locations. Central logging and file-serving
can also consolidate your challenge to a smaller locus, where
you might be able to engineer a more robust solution.

Linux devel mailing lists are a fountain of info on fs reliability
and repair. Also, Toby Oetiker has good tips on torturing fs's using 
the shell to help establish reliability.

HTH,
-- 
Charles

_______________________________________________
Tech mailing list
Tech@lists.lopsa.org
https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to