xfs
[Top] [All Lists]

Re: IDE write cache and journaling file systems

To: Jim Buzbee <James.Buzbee@xxxxxxxxxxxx>
Subject: Re: IDE write cache and journaling file systems
From: Simon Matter <simon.matter@xxxxxxxxxxxxxxxx>
Date: Fri, 19 Apr 2002 08:31:47 +0200
>received: from mobile.sauter-bc.com (unknown [10.1.6.21]) by basel1.sauter-bc.com (Postfix) with ESMTP id 6D73257306; Fri, 19 Apr 2002 08:31:48 +0200 (CEST)
Cc: XFS List <linux-xfs@xxxxxxxxxxx>
Organization: Sauter AG, Basel
References: <3CBEE618.B220A393@xxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
Jim Buzbee schrieb:
> 
> We've got a bit of an issue. From conversations on this list over the
> last few months, it appears as if enabling the write cache on an IDE
> drive is a "bad thing" when using a journaling file system such as XFS.
> But, when talking to drive manufacturers, we are told that if the write
> cache is disabled, the life of the drive is substantially reduced.  This
> puts us in a bit of a hard place. We have little choice but to turn the
> write cache on.

Jim,

It's obvious that lifetime is reduced with write caching off. Just
listen to the drive and compare between w/cache on / off. What helps
here is that Linux uses memory for caching effectively.

Some weeks ago I brought an issue to this list. I got zero filled files
after _clean_ reboots and I thought it was the write cache of the IDE
drives not flushing correctly. In fact I got bitten by the 'remount
readonly bug' and it had nothing to do with the write chache being
turned on.

> 
> In our application, (consumer set top box) we cannot always cleanly shut
> down the system.  The consumer rightly expects to just unplug the box
> when he wants/needs to.  I'm not terribly concerned about losing a bit
> of data in such a case. I'm worried about file system corruption that
> would turn the box into an expensive door stop. My own testing so far
> has not shown any catastrophic failures, but if we have a million boxes
> in the field, issues could start showing up.
> 
> The drive manufactures have recommended inserting IDE cache flushes at
> critical sections of the code. I'm hesitant to muck with XFS internals,
> and adding flushes in our user-space code would not cover all cases.

Cache flushing does only work if the drive honours the flushing request
correctly. What I recommend is trying to use drives which:
- really flush cache when requested to do so.
- flush cache on power failure.

The latter is quite important because comsumers will just pull power to
turn it off. I don't know whether this feature is available with IDE
drives but I think it should be possible.

-Simon

> 
> This has to be a common problem.  Does anyone have any strategies or
> words of wisdom?
> 
> Jim Buzbee,
> Echostar Technologies



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