xfs
[Top] [All Lists]

Re: write cache on 3ware controller and XFS

To: Christophe Zwecker <doc@xxxxxxxxxx>
Subject: Re: write cache on 3ware controller and XFS
From: "Bryan J. Smith" <b.j.smith@xxxxxxxx>
Date: Tue, 05 Nov 2002 16:22:50 -0500 (EST)
Cc: Seth Mos <knuffie@xxxxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <1036531054.20194.10.camel@xxxxxxxxxxxxxxxx>
References: <4.3.2.7.2.20021105215943.03941310@xxxxxxxxxxxxx> <1036531054.20194.10.camel@xxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: IMP/PHP IMAP webmail program 2.2.7
Quoting Christophe Zwecker <doc@xxxxxxxxxx>:
> 256 MB, I dont think it reaches 256mb filling when it stalls

vmstat would prove/disprove that.

> yeah, with write cache enabled its alot faster. its not the network
> thats slow, just the machine with the controller refuses to take more
> data

I'm surprised you're getting corruptions.  Again, I've only seen two cases with
RAID cards where I have issues.

  1.  firmware incompatible with driver
  2.  AMI BIOS and IRQ sharing (nasty)

> 5x 120GB IBM 7200RPM IDE disks, RAID5 on an EVMS/LVM Volume

Okay, now I'm confused.  Why do you need to use EVMS/LVM with a 3Ware card?

-- 
Bryan J. Smith, E.I.            Contact Info:  http://thebs.org
A+/i-Net+/Linux+/Network+/Server+ CCNA CIWA CNA SCSA/SCWSE/SCNA
---------------------------------------------------------------
'In the new commercials for MSN 8, a guy dressed in a Microsoft
 Butterfly costume drops out of a "cocoon" ... nobody at
 Microsoft ... knows that a butterfly doesn't emerge from a
 cocoon, but a chrysalis -- moths come from cocoons ... "this
 should not be all that surprising since Microsoft seems to
 have problems recognizing bugs ..."' -- Cringely



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