Mike Burger wrote:
> I was thinking of any chance of stability, with regard to the file
> system, itself. Having just lived through an ext2 meltdown on my
> primary server (I'm still picking up the pieces), I'm speaking
> from experience.
Really? Hmmm, been using Ext2 on production systems since 1995,
Ext3 since mid-2000. The most I've ever had was a fsck that output
MAXINT length filesizes that had to be cleared with "debugfs" -- but
that was due to a _physical_disk_error_. My volume stayed intact
with exception of ~100 files / ~2MB of data
As an original NT 3.1 beta tester and NT admin through early 1999 (8
years experience), I canNOT say the same about NTFS. A physical
disk error was usually a major pain and it was recommended you get
all the data off you could (ha!) and reformat. In addition, I had 2
non-physical error-related NTFS _total_corruptions_ because of an
_incorrect_journal_read_. I.e., I find NTFS is a little "too
aggressive" in going to the journal whereas filesystems like Ext3
will _always_ drop down to a full "Ext2 fsck" if it detects anything
remotely incorrect.
I haven't worked with XFS long enough to see it take any issue after
an improper shutdown, log/journal read.
-- TheBS
--
Bryan "TheBS" Smith chat:thebs413 @AOL/MSN/Yahoo
Engineer mailto:b.j.smith@xxxxxxxx,thebs@xxxxxxxxxxx
*********************************************************
"Never apply a Star Trek solution to a Babylon 5 problem"
-- Nicholas C. Weaver
|