Thanks guys, I working on a little solution, I'll let you know if it works.
miles
----------------------------------------------------------------------------------------------
-------------------
Miles Purdy
Miles Purdy
System Manager
Farm Income Programs Directorate
Winnipeg, MB, CA
purdym AT fipd.gc DOT ca
ph: (204) 984-1602 fax: (204) 983-7557
"If you hold a UNIX shell up to your ear, can you hear the C?"
-------------------------------------------------------------------------------------------------
----------------------
>>> jmk AT EMAIL.UNC DOT EDU 29-Nov-01 12:50:55 PM >>>
>>> jmk AT EMAIL.UNC DOT EDU 29-Nov-01 12:50:55 PM >>>
You got it! If a volume fails over to another node and is not moved back it
will be backed up as if it was the first time. We try to make sure that the
larger volumes stay on their home node to avoid such duplication. It's a pain
in the youknowwhere!
You can see the arrangement from the Web client. In our case the cluster nodes
all have the same TSM nodename in the dsm.opt but if you go to the restore
screen all the volumes are identified by NW node name, not the TSM node name.
Of course I guess you could (given the resources) go ahead and have all the
volumes backed up from
all the nodes but how ugly is that?
As is so often the case the tie-in between Netware and TSM is not quite soup
yet. As Seth has mentioned a couple of times Novell is supposed to have cluster
aware nlms for TSM in 6.0 or sometime next year (whichever comes first, I
thought 6 was out in Beta already).
Miles Purdy wrote:
> Hi folks, I could really use some help answering this question:
>
> With specific regards to #4, the Novell servers are set up (so I'm told I
> don't administer them) in a cluster with many 'shared' volumes, and of course
> only one node has a volume 'online' at a time. But during a failover
> situation a node may have a volume mounted that it does not normally.
>
> For example servers S1 and S2 and volumes V1 and V2 - S1 normally has V1 and
> S2 normally has V2. Now the problem is that during a failure over from S1 to
> S2, V1 will not be mounted on S2. And if I run a backup of S2 it will backup
> V1, which not really what I want. S2 will backup *all of* V1, instead of
> performing an incremental.
>
> The 'vritualnodename' parameter, according to the documentation, seems to
> work only with restores. No?
>
> The question is:
> So if V1 normally gets backed up on S1, and V1 moves to S2, how do I perform
> an incremental backup of V1 on S2?
>
> Thanks Miles
>
> ----------------------------------------------------------------------------------------------
> Miles Purdy
> Miles Purdy
> System Manager
> Farm Income Programs Directorate
> Winnipeg, MB, CA
> purdym AT fipd.gc DOT ca
> ph: (204) 984-1602 fax: (204) 983-7557
>
> "If you hold a UNIX shell up to your ear, can you hear the C?"
> -------------------------------------------------------------------------------------------------
>
>
> >>> ha7 AT DAIMLERCHRYSLER DOT COM 27-Nov-01 1:44:18 AM >>>
> Some hints ( Novell 5 )
> 1) User with full administrator rights
> 2) entry in dsm.opt
> DOMAIN ALL-LOCAL NDS
> 4) nothing special known to us - we are running a cluster, backup each node
> as a seperate node ( and use virtualnodename if necessary )
>
> hope this helps a little
>
> herfried
>
> Miles Purdy <PURDYM AT FIPD.GC DOT CA>@VM.MARIST.EDU> on 26.11.2001 19:28:03
>
> Please respond to "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
>
> Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
>
> To: ADSM-L AT VM.MARIST DOT EDU
> cc:
>
> Subject: Experienced Admin needs help with Novell
>
> Hi folks,
>
> I currently administer an AIX TSM server and clients, now I need to backup
> Novell servers. I would appreciate some help with some general questions:
>
> -Novell seems to need a user and password to run the scheduler as, what do
> people use?
> -How do I backup up NDS?
> -Any special excludes?
> -What's the best way to handle nodes when they are in a cluster?
>
> Any other gottcha's? I did fix the TSA.
>
> Thanks Miles
>
> ----------------------------------------------------------------------------------------------
>
>
> Miles Purdy
> System Manager
> Farm Income Programs Directorate
> Winnipeg, MB, CA
> purdym AT fipd.gc DOT ca
> ph: (204) 984-1602 fax: (204) 983-7557
>
> "If you hold a UNIX shell up to your ear, can you hear the C?"
> -------------------------------------------------------------------------------------------------
>
>
> ****************************************************************************
>
> The information contained in this transmission, which may be
> confidential and proprietary, is only for the intended recipients.
> Unauthorized use is strictly prohibited. If you receive this
> transmission in error, please notify me immediately by telephone
> or electronic mail and confirm that you deleted this transmission
> and the reply from your electronic mail system.
> ****************************************************************************
--
Jim Kirkman
Jim Kirkman
AIS - Systems
UNC-Chapel Hill
966-5884
|
