SUMMARY: 6120 Disk Array-> Remove and Replace Disk, does not enable

From: Tim Chipman <tchipman_at_gmail.com>
Date: Tue May 15 2007 - 08:07:28 EDT
Hi all, just a summary (a bit late but better than none for the
universe at large I believe)

Big thanks goes out to Alan -- a local person who helped me out.  Alas
I had no replies from the list.  I guess these 6120 units are not in
heavy use (at least by people who don't have support contracts on
them? :-)

Basic Scoop is 2-fold for my problem,

(1) as per Alan's suggestion, it seems there were bugs in early (ie,
the not-latest) firmware for the array which resulted in failure to
consistently pick up when a new drive was installed.  The fix for this
was to download the newest firmware (released in 2007) for all
components in the array (main controller, fibre controllers and the
drives themselves).  After applying these updates to the array it
still would not let me use the 3rd party (non-Sun FCAL Drive, lacking
sun firmware but otherwise physically identical drive) -- same part
from seagate as the failed drive - nor could I flash it up to the sun
firmware (no big surprise really I guess)

(2) I was able to source a pair of sun-part drives (good used working
pulls) on Ebay for a bargain price  (same seagate part as my failed
drive, but also with Sun firmware).  After popping one of these into
the array, I was able to update the firmware of the new drive to make
it consistent with all other drives in the array. Following this, I
was able to manually enable the drive and then a reconstruction began
immediately on the raid volume, taking a significant chunk of time
(overnight) to rebuild - but completing successfully.  (the rebuild
time reminds me .. this is basically a souped-up T3 disk array, ie,
not exactly cutting edge performance to put it mildly. Sigh.)

So.  I'm all finished with this madness for now, it seems.  The joys
of non-contract gear.

---Tim Chipman


===============
ORIGINAL QUESTION FOLLOWS BELOW (with console array text truncated
this time, which was pasted / present in my original posting)

Hi All,

Sorry to bother the list with trivial question, but the sun manuals
are absurdly opaque on this topic ("it happens automatically in the
default config, yay")

I've inherited management of a 4800 with a couple of direct-attached
6120 arrays (everything is out of support so I'm working with access
only to "free sunsolve", and no tech support from sun).  One of the
arrays has a disk that went bad - not a problem since we're running
raid 5.

I've got a cold-spare part which is the identical part as the failed
drive (Seagate st373307fc, 73gig FCAL disk - although my cold spare is
a generic seagate part, NOT a sun part)

First question, I suppose, is - do I have any reason to this this
should work with a non-sun part? or will firmware of the disk make it
incompatible from the get-go? (hopefully not?!)

Second question, assuming I have any right to be doing this with a
generic disk:

When I remove the bad disk, and install the replacement part, the new
disk spins up fine (power LED on the new disk lights up persistently)
and I can see from the command-line-admin-interface that the device is
detected.  However, it does not come online, nor does it rebuild the
raid.

A friendly kick in the right direction would be massively appreciated.
(i.e., what command syntax is expected to force the new disk to be
flagged as online and then permit rebuild of the raid volume?!)
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Received on Tue May 15 08:07:48 2007

This archive was generated by hypermail 2.1.8 : Thu Mar 03 2016 - 06:44:05 EST