[Linux-cluster] Re: Linux-cluster Digest, Vol 35, Issue 18

[Alan Wood <chekov ucla edu>]

Holger,

we ran into that problem and went back and forth with dell for literally 
months as they replaces every component in the 220, our Perc cards, cables, 
etc.  we were never able to solve the problem and I ended up buying a 
couple of iSCSI enclosures instead, which I've been very happy with.  Much 
of the consensus from experts I've talked to is to avoid using shared SCSI 
buses for concurrent access, at least with the megaraid driver.  I have not 
had any problems with the PV enclosure since shifting it to a single-system 
connection.

wish I could tell you something more positive.  In the end I was not able 
to correlate the crashses with anything (load, calendar, processes, users, 
or settings).  Sometimes it would go a month without crashing sometimes it 
would crash many times in a week. 
-alan

> ------------------------------
>
> Date: Wed, 14 Mar 2007 15:55:32 +0100
> From: "Holger L. Ratzel" <holger ratzel she net>
> Subject: [Linux-cluster] Problem with shared SCSI-Storage
> To: linux-cluster redhat com
> Cc: "Stempel, Steffen" <Steffen Stempel she net>
> Message-ID: <200703141555 32868 holger ratzel she net>
> Content-Type: text/plain;	charset="us-ascii"
>
> Hi All,
>
> I've a problem with a RHEL4 cluster running on two Dell PowerEdge 1850 and an
> Dell PowerVault PV220S. After running for a while I get errors from the
> megaraid driver in /var/log/messages saying that the device was "offlined"
> and the service gets relocated to the other node while the original node
> reboots. The relevant parts from /var/log/messages can be downloaded from
> http://www.she.net/hra/web01-messages.txt
>
> Dell has checked the system and all firmwares have been updated, so it seems
> not to be a hardware problem.
>
> Please mail me if you have any questions or need further information.
>
> I would appreciate any suggestions.
>
> Many thanks an best regards,
>
> 	Holger Ratzel