Re: SCSI problems with a 2650

Russell Stuart <[email protected]> 04 Aug 2003 11:28:33 +1000
Newsgroups gmane.linux.drivers.aacraid.devel,gmane.linux.hardware.dell.poweredge
Organization Lube Mobile
Message-ID <1059960513.15637.63.camel@ras>
On Sat, 2003-07-26 at 03:42, James Bourne wrote:
> > I have the problem only on 2650 running dual 2.0Ghz Xeon, and not on 
> > 2650 running dual 2.4Ghz Xeon, while I have exactly the same Linux 
> > flavor, with exactly the same drivers.
> 
> Possibly due to not high enough CPU usage and I/O to the disks.  We have
> seen the problem on both 2650's and 1650's.  Currently I'm running a test
> with write caching off.  It has been running now for about 10 days and I
> have seen several command timeouts but to this point the
> system has not completely cratered.
> 
> Regards
> James Bourne

For what it is worth, I have been running the test that failed earlier
with write caching off.  It has been running for 9 days, and has worked
perfectly.  No dead machines, no SCSI timeouts.  Configuration tested:

Test:   2650, kernel 2.4.20-18.7.  Tested with SMP+SMT off, write
        caching off in controller.
Result: Machine works perfectly.  No SCSI entries in /var/log/messages.

This configuration was failing within 24 hours when write caching was
on.

Following other pointers on here about SCSI protocol errors, I am going
to try and upgrade the firmware in my Segate drives.  (I am already
running the latest firmware for the embedded Adaptec RAID controller). 
Getting hold of the right firmware drive looks to be the biggest
challenge.

_______________________________________________
Linux-aacraid-devel mailing list
[email protected]
http://lists.us.dell.com/mailman/listinfo/linux-aacraid-devel
Please read the FAQ at http://lists.us.dell.com/faq or search the list archives at http://lists.us.dell.com/htdig/