I'm using a Promise VTJ610SS SAS enclosure populated with 16 Samsung
HD203WI 2TB disks. About every 20 minutes, the hard error count
reported by iostat increments by exactly 9 for each disk in the
external enclosure. An identical Samsung disk (c7t27d0) connected to
an internal ICH10R SATA controller shows zero hard errors. Below is
the output from iostat -en:

  ---- errors ---
  s/w h/w trn tot device
    0   9   0   9 c8t0d0
    0   0   0   0 c7t8d0
    0   0   0   0 c7t9d0
    0  90   0  90 c7t11d0
    0  90   0  90 c7t12d0
    0  90   0  90 c7t13d0
    0  90   0  90 c7t14d0
    0  90   0  90 c7t15d0
    0  90   0  90 c7t16d0
    0  90   0  90 c7t17d0
    0  90   0  90 c7t18d0
    0  90   0  90 c7t19d0
    0  90   0  90 c7t20d0
    0  90   0  90 c7t21d0
    0  90   0  90 c7t22d0
    0  90   0  90 c7t23d0
    0  90   0  90 c7t24d0
    0  90   0  90 c7t25d0
    0  90   0  90 c7t26d0
    0   0   0   0 c7t27d0

There are no errors in the syslog to explain why the hard error counts
are increasing. The enclosure is connected to a Nehalem server running
OpenSolaris 2009.06 with an LSI1068E based HBA (SAS3442E-R). I have
another server with identical hardware (except WD RE3 disks instead of
Samsung). and it does not show any iostat errors at all.

Incidentally, I am also seeing the infamous mpt driver timeout lockups
on this system. I would appreciate help in eliminating these hard
errors and solving the mpt driver lockup. I wonder

Thanks,

Daniel Bakken
_______________________________________________
storage-discuss mailing list
storage-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to