Oops, sorry, in my last message I incorrectly stated I used the patch
for 2.2.10, I meant 2.2.11 of course,
-Malte
--
570 Professoren glauben, daß Atomunfälle auch künftig
immer nur woanders passieren
of course I have a level 1 raid (mirroring)
NOT level 0 (striping)
Daniel
On Fri, 22 Oct 1999, Daniel Wirth wrote:
> Date: Fri, 22 Oct 1999 11:04:37 +0200 (MEST)
> From: Daniel Wirth <[EMAIL PROTECTED]>
> Reply-To: [EMAIL PROTECTED]
> To: [EMAIL PROTECTED]
> Cc: [EMAIL PROTECTED]
> Subject: Re
And lo, Eyal Lebedinsky saith unto me:
>
> Jon Lewis wrote:
> >
> > On Sat, 26 Dec 1998, Eyal Lebedinsky wrote:
> >
> > > I think the FAST-5 was negotiated as the best the drive will do. Here is
> > > how another disk comes up:
In your kernel configuration, be sure that you've specified someth
Eyal Lebedinsky <[EMAIL PROTECTED]> wrote:
> I really think the disks are just stubborn.
Yes. Moreover they might handle disconnection unefficiently, which will
kill your RAID0 performance completely.
Jon Lewis wrote:
>
> On Sat, 26 Dec 1998, Eyal Lebedinsky wrote:
>
> > I think the FAST-5 was negotiated as the best the drive will do. Here is
> > how another disk comes up:
>
> Are the old disks SCSI-1 or SCSI-2? If you don't have important data on
> them, you might try forcing 10MB/s and se
On Sat, 26 Dec 1998, Eyal Lebedinsky wrote:
> Good point. Here is what I get (with vmstat, MB/sec)
> 1 disk1400
> 2 disks 2400
> 3 disks 3000
> 4 disks 3200
> then no more improvement.
> Direct-Access ANSI SCSI revisi
[EMAIL PROTECTED] wrote:
>Eyal Lebedinsky wrote:
>> the disks seem to be slow on their own
>> (hdparm says 1.36 MB/sec) and I expected some speedup when striped
>> (raid0). I got 2.8 MB/sec,
> Perhaps check that you can get more than 2.8MB/sec across the bus when running the
>discs as six/n separ
"David Harris" wrote:
>Eyal Lebedinsky wrote:
>> the disks seem to be slow on their own
>> (hdparm says 1.36 MB/sec) and I expected some speedup when striped
>> (raid0). I got 2.8 MB/sec,
>
>You mention that you are trying to achieve a read speed boost from
>"striping", but then mentioned that you
