| To: | Andrew Klaassen <ak@xxxxxxx> |
|---|---|
| Subject: | Re: XFS and RAID5 |
| From: | ctooley@xxxxxxxx |
| Date: | Mon, 18 Jun 2001 14:03:02 -0500 |
| Cc: | linux-xfs@xxxxxxxxxxx |
| Sender: | owner-linux-xfs@xxxxxxxxxxx |
One of the things I noticed with the Promise Ultra66 (may be fixed in Ultra100)
was that if a disk failed, both disks on that controller (master and slave)
start having a load of problems. Plus there is no way to reset just part of the
controller, it's both or nothing. By just using the Master connectors I noticed
that when I did my testing setting the bus that it was on after replacing the
disk (again DON'T DO THIS!!!) was all I had to do to make the disk available.
Chris
Andrew Klaassen <ak@xxxxxxx> on 06/18/2001 12:17:29 PM
To: linux-xfs@xxxxxxxxxxx
cc: (bcc: Chris Tooley/AMOA)
Subject Re: XFS and RAID5
:
On Tue, Jun 19, 2001 at 02:55:26AM +1000,
Robin Humble wrote:
> Contrary to popular belief, using both master and slave only
> gets you a ~5% performance hit.
(?)!
That would be very, very good news. Has anybody else tried
this? Have you load tested the machines and seen smooth
performance degradation, no nasty bottlenecks (or whatever it is
that is supposed to happen when master and slave are both used
in an array)?
(Am I asking these question on the wrong forum? Apologies...)
Andrew Klaassen
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Testing XFS+RAID5, Simon Matter |
|---|---|
| Next by Date: | Re: ACL Support, Martin Maciaszek |
| Previous by Thread: | Re: XFS and RAID5, Ed McKenzie |
| Next by Thread: | using xfsdump with 'file device', Christoph Lukas |
| Indexes: | [Date] [Thread] [Top] [All Lists] |