RE: ASM & split-mirrors
Date: Mon, 4 Mar 2013 02:34:53 +0000
Message-ID: <850425C04FB76547AC0ED12C5F7E0B800F07A2_at_HEIMSEAT021.ahc.root.loc>
Sounds good.
Assuming I do give it the correct set of LUNs to examine, how does it make sense of what it finds? That is, it will locate a bunch of LUNs with the proper signature on the header, but how does it know what is what?
From: Matthew Zito [mailto:matt_at_crackpotideas.com]
Sent: Sunday, March 03, 2013 7:48 PM
To: Hubler, Daniel
Cc: Ls Cheng; oracle-l_at_freelists.org
Subject: Re: ASM & split-mirrors
ASM won't care - it only looks at the signature on the header of the disk (assuming your scan path is set to look at the right set of LUNs). You only run into issues when you do something like clone a lun, so the same device appears at both /dev/sda and sdb on the same server, as ASM won't know which LUN is the "real" one.
Matt
On Sun, Mar 3, 2013 at 8:30 PM, Hubler, Daniel <daniel.hubler_at_aurora.org<mailto:daniel.hubler_at_aurora.org>> wrote:
That sounds pretty routine.
Thanks.
One thing is bothering me. . . . .
When the LUNs are established on the TARGET side, they will have different names/identifiers than on the source side.
That is, /dev/diska on the source side will become /dev/diskx on the target.
How does that get reconciled within ASM?
-- http://www.freelists.org/webpage/oracle-lReceived on Mon Mar 04 2013 - 03:34:53 CET