HP3000-L Archives

March 2004, Week 2

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Craig Lalley <[log in to unmask]>
Reply To:
Craig Lalley <[log in to unmask]>
Date:
Wed, 10 Mar 2004 08:57:05 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (58 lines)
OK John, I will give it a shot.

  Regarding multiple volume sets.

> Is it a good thing?  Does it complicate system management tasks? If so, by
> how much?  Also, how would we get there from here, are we looking at a
> reload? Can you point me toward documentation that might list the steps
> required?

Multiple volume sets provided some fault tolerance when mirroring was not
available.  It also provides additional instances of the XM manager, one per
volume set.  This can be a good thing if the system is write intensive.
Remember the XM is a write cache.  So your mileage may vary depending on the
type of transactions involved.

>
> Second question. Currently, all the disk drives are set up with the same
> device class name: disc.  We'd like to split drives 2-13 into a separate
> device class name (e.g. datadisc, or something similar), so we could exclude
> LDEV1 when we create files  (eg :build blah;dev=datadisc). Can I attach a
> second device class name to the desired disk drives without a reload?  Can I
> do it while users are on the system, or do I need exclusive access?

Yes, IOCONFIG or SYSGEN with DOIONOW should work.  ADDCLASS is the command
(IIRC), just add it to the currently existing drives.

>
> Third (and last) question.  Like I said above, we have 13 LUNs spread across
> 2 Nike Mod 20s (actually, 13 sets of mirrored pair drives with a couple of
> hot spares thrown in for good measure). The Nikes go off HP maintenance at
> the end of June. We have been unsuccessful in getting information from HP as
> to whether we can pay extra and keep those arrays a special support
> contract, or if there is a reasonably-priced migration path for those arrays
> (we know about the VA-7110; no one here considers it reasonably-priced...).
>  I'd be interested in hearing ideas about what we should do after June. Is
> there a reasonable pool of Mod 20 parts in the after market?  What about the
> individual drives (we have both 9 and 4 gig drives)?  We are in the
> migration boat, and don't expect to stay on the 3000 much beyond 2007, so it
> wouldn't be long-term.

My recommendation would be to by several spare drives, both 9 and 4 gig.  Then
by a spare Mod 20 cabinet, total cost would be considerably less than the
support contract.  VA-7100 can be reasonably priced, you would need to shop
around, but the would also give you much higher performance.

Hope that helps.

-Craig


__________________________________
Do you Yahoo!?
Yahoo! Search - Find what you’re looking for faster
http://search.yahoo.com

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

ATOM RSS1 RSS2