I guess in your case the name persistence was turned on and hence DMP
honoured it and used the name it had chosen previously. In order to set
the names right, you can execute the following VxVM 5.0MP3 command on
server B to get the names aligned with server A

        vxddladm assign names

Please note the names displayed by "vxdisk list" is just a
representation for the DMP device

With regards
Hari
-----Original Message-----
From: veritas-vx-boun...@mailman.eng.auburn.edu
[mailto:veritas-vx-boun...@mailman.eng.auburn.edu] On Behalf Of
veritas-vx-requ...@mailman.eng.auburn.edu
Sent: Tuesday, January 06, 2009 11:30 PM
To: veritas-vx@mailman.eng.auburn.edu
Subject: Veritas-vx Digest, Vol 33, Issue 2

Send Veritas-vx mailing list submissions to
        veritas-vx@mailman.eng.auburn.edu

To subscribe or unsubscribe via the World Wide Web, visit
        http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx
or, via email, send a message with subject or body 'help' to
        veritas-vx-requ...@mailman.eng.auburn.edu

You can reach the person managing the list at
        veritas-vx-ow...@mailman.eng.auburn.edu

When replying, please edit your Subject line so it is more specific than
"Re: Contents of Veritas-vx digest..."


Today's Topics:

   1. Re: Veritas DMP question. (Ronald S Karr)


----------------------------------------------------------------------

Message: 1
Date: Mon, 5 Jan 2009 23:36:46 -0800
From: Ronald S Karr <s...@spencerkarr.com>
Subject: Re: [Veritas-vx] Veritas DMP question.
To: Jason Morgan <ripples2...@hotmail.co.uk>
Cc: veritas-vx@mailman.eng.auburn.edu
Message-ID: <ebe9f017-6d8e-4bf5-9f2a-6bd31dec1...@spencerkarr.com>
Content-Type: text/plain; charset="us-ascii"

The algorithm was modified to sort the names of the paths and to (where
possible) use the first name.  There is a flag somewhere to make the
names persistent.  Most people seemed less confused by the
non-persistent sorting algorithm than by having stale, no-longer-used
paths used for the DMP node names.  On some operating systems (though
not Linux), this is also a bit more likely to result in consistent node
names in a cluster.

The enclosure-based naming schemes have also been gradually improving,
in attempting to name the DMP nodes more based on what is out there on
the SAN than based on pure (and mostly arbitrary) internal OS device
discovery order.
-- 
     Ron Karr

On Dec 24, 2008, at 7:31 AM, Jason Morgan wrote:
> Hi
>
> I have built/rebuilt two servers today with RedHat 4U6. Both servers 
> have the same number/size LUNS presented to each.
>
> Each server has 4 paths to the disks.
>
> Server A was a brand new build. After Volume Manager 5.1 was 
> installed/disks initialized and added to Veritas, the device name was 
> /dev/sdd.
>
> vxdisk list sdd showed four paths as I expected and the correct LUN 
> number.
>
> e.g /dev/sdd
>      /dev/sdas
>      /dev/sdba
>     /dev/sdca
>
> Server B has the LUNS removed and represented . A reboot of the server

> was done after this and the same Volumen Manager procedure was used as

> in Server A
>
> However when a vxdisk list is now run , the Physical Device column now

> shows /dev/sdba
>
> and vxdisk list /dev/sdba show the entries below which is correct.
>
> lists as /dev/sdba
>           /dev/sdas
>           /dev/sdba
>           /dev/sdd
>
>
> It looks as though DMP chose a different disk device to put into 
> Veritas. Although functionally this works ok , is there any way to 
> alter the Physical device name so that Server B show the devices as in

> Server B i.e sdd sde etc in vxdisk list. Or is it the fact that the 
> LUNs were removed/represented. It would be a shame to have to rebuild 
> the server to get the Disk names persisitent.
>
> Thanks in advance.
>
> Jas
>
>
> Take your friends with you with Mobile Messenger. Click Here!  
> _______________________________________________
> Veritas-vx maillist  -  Veritas-vx@mailman.eng.auburn.edu 
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://mailman.eng.auburn.edu/pipermail/veritas-vx/attachments/20090105/
0911c972/attachment-0001.htm

------------------------------

_______________________________________________
Veritas-vx maillist  -  Veritas-vx@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx


End of Veritas-vx Digest, Vol 33, Issue 2
*****************************************

_______________________________________________
Veritas-vx maillist  -  Veritas-vx@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx

Reply via email to