[Top] [All Lists]

Re: LBA to File?

To: linux-xfs@xxxxxxxxxxx
Subject: Re: LBA to File?
From: Charles Shannon Hendrix <shannon@xxxxxxxxxxxxx>
Date: Fri, 7 Mar 2003 20:14:39 -0500
In-reply-to: <1046992493.1388.235.camel@averell>
Mail-followup-to: linux-xfs@xxxxxxxxxxx
References: <3E67BFA7.4080601@xxxxxxxxxxxx> <20030306221631.GA6770@xxxxxxxx> <3E67CB02.2030600@xxxxxxxxxxxx> <1046990328.439.273.camel@xxxxxxxxxxxxxxxxxxxxxx> <3E67D0CE.3080902@xxxxxxxxxxxx> <1046992493.1388.235.camel@averell>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.4i
On Fri, Mar 07, 2003 at 12:14:53AM +0100, Andi Kleen wrote:

> It was long suspected that journaling FS are bad for cheap IDE disks,
> but you're probably the first to really verify it using an IDE analyzer.
> There were also some informal experiences on that collected by the
> reiserfs people. Thanks for verify this.

I've thought about this too, as I once found damage in superblocks
even on other drives, across about 5 different IDE drives that had
"gone bad".  I found error patterns, and it appeard to me it was occuring
in superblocks and in one case, a journal.

Does this mean that these areas are also wearing out even good disks,
just not as fast?

I know over the last 10 years, I've replaced IDE drives at a rate several
times higher than SCSI drives, even for "good" IDE drives.  They don't
seem to last very well.

Seems like the high-end Seagates and IBMs aren't too bad so far, but
I've not run the IDE versions long enough to see problems yet.

UNIX/Perl/C/Pizza____________________s h a n n o n@wido !SPAM maker.com

<Prev in Thread] Current Thread [Next in Thread>