Markus Kovero wrote:
> Hi, we have currently setup that involves latest Opensolaris  on Dell server
> connected to 16bay sas/sata JBOD with sas x4 (sff-8088 - sff-8470) cabling. 
> Everything seemed fine at first but then we had random disk timeouts when
> doing heavy writes to disks, we used raidz2 and split striped raidz2
> configurations. Problems went away after we forced disks to 1.5Gbps with
> jumpers, disks on system are WD 1TB black edition. Has anybody encountered
> similar issues? Are these JBOD / disk firmware related? Any more feasible
> workarounds than adding a jumper to every disk and settling with lower
> performance?

Markus,
Those WD drives are desktop models, not enterprise drives.  Desktop drives have
error-correcting routines that can sometimes cause the drive to take longer to
respond than the controller will wait, so you see timeouts.  I just had a bad
experience with desktop drives in a large storage server, so I would recommend
running enterprise SATA drives, such as the Seagate Barracuda ES.2 or WD "RE"
(RAID Edition) drives.

We've had good experience with the ES.2 1TB drives.  Yesterday we rebuilt our
storage server with 2TB WD RE4 drives.  The RE drives have a feature called
Time-Limited Error Recovery (TLER) which should reduce or eliminate any timeout
issues.  The Seagate drives have something similar but they don't have a catchy
acronym for it.

It's too early to tell yet on the reliability front for the 2TB drives, but they
seem to perform quite well.

Hope this helps,
Eric

-- 
Eric Sproul
Lead Site Reliability Engineer
OmniTI Computer Consulting, Inc.
Web Applications & Internet Architectures
http://omniti.com
P: +1.443.325.1357 x207   F: +1.410.872.4911
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to