On Mon, 16 Aug 2010 09:13:54 -0700
[email protected] (Randal L. Schwartz) wrote:
> For single spindle failure, the disks are on RAID10, so I'm not
> terribly worried. (Far better than what I used to have, which was a
> nightly rsync to a second spindle.)
Don't discount that nightly rsync to a second (normally unmounted)
disk. That policy saved us one time when a junior admin added a new
account to an internal Solaris 2.5.1 box by using
echo 'user:uid:etc.etc.etc.' > /etc/passwd
instead of using the provided tools. Yeah, the password file suddenly
became one line long. We were able to boot from the mirror and restore
the previous day's password file.
At least he has a good story to tell during an interview when asked
to describe a mistake he made and what he learned from it. :-)
> For fumble-fingers, I have ZFS snapshots...
Snapshots are a wonderful thing. Not available back then, but would
also have easily resolved the problem.
-jan-
--
Jan L. Peterson
http://www.peterson-tech.com/~jlp/
_______________________________________________
Tech mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
http://lopsa.org/
