25 Sep
2008
25 Sep
'08
3:56 p.m.
On Thu, Sep 25, 2008 at 05:31:52PM -0500, Jim Leonard wrote:
Alexander Schreiber wrote:
Most, not all. Running e2fsck against a device that is read-only and
misses its second half causes e2fsck to die painfully - BTST.
Besides, if the data was important, you have current backups, right? And
if it is just the machine setup, you have the package list and a backup
of the config files, correct? ;-)
Regards,
Alex.
--
"Opportunity is missed by most people because it is dressed in overalls and
looks like work." -- Thomas A. Edison
On Thu, Sep 25, 2008 at 11:07:55AM -0500, Jim
Leonard wrote:
You're operating under the impression that e2fsck can fix all failures. Alexander Schreiber wrote:
That's what fsck is for. Seriously. What most journaled file systems gain
you is that you don't need to spend 20 min in fsck after a powerfailure,
just less than a minute in journal rollback. And besides: both ext2
and, if needed, e2fsck are extremly robust. - do not use a journalling FS on the CF
Just
because the device is solid-state doesn't mean you should ignore
journaling. The host computer can still abort halfway through a write
and bork the filesystem.