HP3000-L Archives

March 2003, Week 4

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:
John Lee <[log in to unmask]>
Reply To:
Date:
Tue, 25 Mar 2003 10:04:18 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (98 lines)
Tim is correct.  There is a lot of confusion about this controller issue,
but Tim explains it well.

And I'll add an opinion here...as inexpensively as a Model 10 or 20 disk
array can be integrated today, if you aren't RAIDing your disk in some
form, you are making a bad decision.  Disk failure is the number one cause
of downtime, and you can eliminate it with one of these.  And they are easy
to integrate.

John Lee
Vaske Computer Solutions

At 04:31 PM 3/24/03 -0800, Atwood, Tim (DVM) wrote:
>I will jump in and say "no". At least not in the same sense as HP-UX. MPE
>does not recognize two different controllers pointing to the same set of
>LUN's. MPE does not know how to handle accessing the same data through two
>different paths.
>
>Prior to the release of 7.5, I had been told by several sources MPE 7.5 was
>going to be able to do this. However, so far as I have been able to tell
>this was not in 7.5 when it was released. It is possible I am missing
>something or it is still to come in 7.5 Express 1. If anyone has information
>on 7.5 and multiple controllers, I would love to hear it.
>
>I suppose one could configure one set of LUN's on one controller and another
>set of LUN's on the second controller. So if you have two interface cards
>you could configure the path of each card to each controller. Essentially
>creating two sets of drives. I have never tried this and I do not know how
>it would work with mirroring.
>
>But as far as I know any particular piece of data (disc address) would still
>only be accessed through one path and never the other path. For any one disc
>IO MPE can not do an "either/or" scenario between two controller cards. It
>would still only service the request through the controller for the LUN
>where the data resided.
>
>-----Original Message-----
>From: Craig Lalley [mailto:[log in to unmask]]
>Sent: Monday, March 24, 2003 3:52 PM
>To: [log in to unmask]
>Subject: Re: Model 20 Disk array
>
>
>Gilles,
>Thanks for the info.  But now I am getting conflicting information. Would
>anyone else care to comment?
> Gilles Schipper <[log in to unmask]> wrote:
>
>>Yes, mod20's can be INSTEAD OF software disk mirroring on the HP3000.
>>
>>Better still, you can use mod20's to mirror the MPEXL_SYSTEM_VOLUME_SET -
>>something which cannot be accomplished with MIRROR/iX software.
>
>Great, that is what I would want to do.
>
>>Also, the mod20 permits dual-controller access to the HP3000 - a feature
>>NOT available with the far more expensive Autoraid 12H.
>
>Is this true?  I have heard about 3 yes and 3 no's.
>
>>Each mod20 supports a maximum of 8 luns - the same limit as the
>Autoraid12H.
>
>8 luns, but 20 mirorred disks, hmmm doesn't compute.  So are you saying I
>can only have a maximum of 8 disks mirrored on two channels?
>
>>In a nutshell, I highly recommend the nike mod20 disk array.
>>
>>I suggest dual Storage Processors (SP's) and maximum memory (which,
>>officially is 64mb per SP, but 128mb per SP works just fine.
>
>Has anyone else done this?  This could be the reason why Larry couldn't get
>it to work.
>
>Thanks in advance.
>
>-Craig
>
>
>
>
>
>
>---------------------------------
>Do you Yahoo!?
>Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop!
>
>* To join/leave the list, search archives, change list settings, *
>* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>
>* To join/leave the list, search archives, change list settings, *
>* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>
>

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

ATOM RSS1 RSS2