Storage space partition

[Nair, Murlidharan T]

-----Original Message-----
From: redhat-install-list-bounces at redhat.com
[mailto:redhat-install-list-bounces at redhat.com] On Behalf Of Bob McClure
Jr
Sent: Friday, April 14, 2006 5:32 PM
To: Getting started with Red Hat Linux
Subject: Re: Storage space partition

On Fri, Apr 14, 2006 at 05:20:34PM -0400, Allen, Jack wrote:
> 
> 
> -----Original Message-----
> From: redhat-install-list-bounces at redhat.com
> [mailto:redhat-install-list-bounces at redhat.com] On Behalf Of Bob
McClure Jr
> Sent: Friday, April 14, 2006 4:59 PM
> To: Getting started with Red Hat Linux
> Subject: Re: Storage space partition
> 
> 
> On Fri, Apr 14, 2006 at 04:35:38PM -0400, Allen, Jack wrote:
> > 
> > 
> > 	When using a SAN and there are multiple paths to it, you will
see
> > multiple sdX entries for the true storage. This means you could
access
> > /dev/sda and see the exact same things if you access /dev/sdb as an
> example
> > if the SAN had been configure to allow access to 1 LUN. With EMC's
> > Navisphere and PowerPath installed, there will also be a
/dev/emcpowera.
> > This is what you want to either partition and or make the whole
things a
> > Physical Volume in LVM. When you run the pvcreate command, or many
of the
> > other LVM related command it scans the /dev directory and reads each
block
> > device it find to see if it is part of a Volume Group and/or can be
used
> by
> > LVM. This is why you are getting the duplicate messages and the
message
> > about the cdrom.
> > 
> > 	To keep this from happening you have to edit /etc/lvm/lvm.conf.
> > There are some pretty good comments that explains things. The main
idea is
> > to exclude the /dev/sd* and /dev/cdrom names and only use the
> /dev/emcpower*
> > names.
> 
> Why would you prefer to use /dev/emcpower* rather than /dev/sd*?
> Seems the latter would yield fewer surprises, even if it's exactly
> equivalent.  Just curious.
> @@@@@@@@@@
> Because if you use the /dev/dsb/sda and that path becomes inaccessible
you
> will not be able to use the disk. This is the whole purpose of EMC's
> PowerPath. The /dev/emcpower* are pseudo devices and the software will
> determine which path can be used. The PowerPath software also does
load
> balancing between the path. There can be many more than 2 paths. You
could
> have 2 FC HBAs that each are connected to a FC Switch, then each FC
Switch
> has 2 paths to the SAN, 2 for each SAN SP (Storage Processor). This
would
> give 4 paths to each LUN. So if the SAN has presented 4 LUNs the
system
> would see 16 /dev/sd* entries and the PowerPath software would create
> /dev/emcpowera through /dev/emcpowerd. This would hold true even if
the
> /dev/sd* started at say /dev/sdc.
> @@@@@@@@@

Okay, that makes sense.  I didn't realize the significance of /dev/sdb
and /dev/sdc.

> > 	This is a little brief in the scheme of things, but I hope it
helps.
> > I can try to give more detail if needed.
> > 
> > Talk 2 U later.
> > Jack Allen
> 
> Cheers,
> -- 
> Bob McClure, Jr.
> ===========================================================
> See answers and comments above between @@@@@@@@@@.
> 
> Talk 2 U later.
> Jack Allen

Cheers,
-- 
Bob McClure, Jr.             Bobcat Open Systems, Inc.
bob at bobcatos.com             http://www.bobcatos.com
The best things in life aren't things.

_______________________________________________
Redhat-install-list mailing list
Redhat-install-list at redhat.com
https://www.redhat.com/mailman/listinfo/redhat-install-list
To Unsubscribe Go To ABOVE URL or send a message to:
redhat-install-list-request at redhat.com
Subject: unsubscribe


I finally finished it. I removed the volume like I mentioned before
using pvremove then recreated it using pvcreate and this time used 
pvcreate /dev/emcpowera1  which was the name given to the partition. I
also added the following filter to lvm.conf
filter = [ "a/sda[1-9]$/", "a/sda[1-9][0-5]$/" "r/sd.*/", "a/.*/" ]

I am so relieved. Thank you all for your help and comments. They were
indeed very helpful. Now I am going for a well deserved game of Tennis
!!

Cheers always!!
Murli