Re: Question about Veritas Cluster and filespace name

2002-12-13 Thread Jim Smith
All,

The TSM Backup-Archive client has support for MSCS (Windows), HACMP (AIX)
and NCS (NetWare) clustering solutions.  If you are using a clustering
solution on one of the Intel platforms other then these listed, you will
have these naming issues after a fail-over.

We are looking into something similiar to what Peter is suggesting, and
that is a way to logically manage a set of data on the TSM server which is
sent from different client nodes.  This would be beneficial for clustering
as well as have other applications.  We are also pursuing adding support
for some of these other clustering solutions.

It doesn't hurt to open requirements to Tivoli if you have a particular
cluster solution that you would like to see supported.

Thanks,
Jim Smith
TSM client development

In response to:

You know, there seems to be a similar issue with filespace naming with
the
new Netware cluster aware client.
We're awaiting a server upgrade to pursue it further. Seems as it the
'getservername' API call is at the heart of this.


Yes, it seems TSM is not savey of any cluster soloutions except
Microsoft's.

If they (TSM) would only introduce a new option in the .opt file so one
could override the servername
if for some reason or another one should wish to do so.
It would probably solve the naming problem in most cluster sitiations.

Regards

Peter

J.P. (Jim) Smith
TSM Client Development
[EMAIL PROTECTED]



Re: Question about Veritas Cluster and filespace name

2002-12-12 Thread Peter Bjoern
You know, there seems to be a similar issue with filespace naming with the
new Netware cluster aware client.
We're awaiting a server upgrade to pursue it further. Seems as it the
'getservername' API call is at the heart of this.


Yes, it seems TSM is not savey of any cluster soloutions except
Microsoft's.

If they (TSM) would only introduce a new option in the .opt file so one
could override the servername
if for some reason or another one should wish to do so.
It would probably solve the naming problem in most cluster sitiations.

Regards

Peter



Re: Question about Veritas Cluster and filespace name

2002-12-12 Thread Joshua Bassi
 Yes, it seems TSM is not savey of any cluster soloutions except
Microsoft's.

That an HACMP.


--
Joshua S. Bassi
IBM Certified - AIX 4/5L, SAN, Shark
Tivoli Certified Consultant -ADSM/TSM
eServer Systems Expert -pSeries HACMP

AIX, HACMP, Storage, TSM Consultant
Cell (831) 595-3962
[EMAIL PROTECTED]

-Original Message-
From: ADSM: Dist Stor Manager [mailto:[EMAIL PROTECTED]] On Behalf Of
Peter Bjoern
Sent: Thursday, December 12, 2002 6:07 AM
To: [EMAIL PROTECTED]
Subject: Re: Question about Veritas Cluster and filespace name


You know, there seems to be a similar issue with filespace naming with
the
new Netware cluster aware client.
We're awaiting a server upgrade to pursue it further. Seems as it the
'getservername' API call is at the heart of this.


Yes, it seems TSM is not savey of any cluster soloutions except
Microsoft's.

If they (TSM) would only introduce a new option in the .opt file so one
could override the servername if for some reason or another one should
wish to do so. It would probably solve the naming problem in most
cluster sitiations.

Regards

Peter



Question about Veritas Cluster and filespace name

2002-12-11 Thread Peter Bjoern
Hi

We have a situation with a Veritas cluster installed on two Win2000
servers.
The physical servers are called J2P-CLUSTER101 and J2P-CLUSTER102.
There is a virtual server defined with the name WEB1.

Each physical server has it's own C: drive which is backed up via a
schedule
that specifies 'C:' and 'systemobject' on the domain statement in the
option file.
The C: filespaces are given the names \\J2P-CLUSTER101\C$ and
\\J2P-CLUSTER102\C$ respectively.  This is all OK.

There is a shared drive X: which is associated with the WEB1 resource and
moves with WEB1 between the two physical servers so that it is available
either on J2P-CLUSTER101 or J2P-CLUSTER102.

X: is backed up from a virtual node defined in TSM called WEB1 and via a
separate TSM scheduler
service with an option file that  specifies X: on the domain statement.

If this had been a Microsoft Cluster installation, we would have specified
'CLUSTERNODE YES' in the optionfiles to make TSM name the X: filespace
with the virtual clustername (WEB1) so that it would have been named
\\WEB1\X$ regardless of which physical server it was backed up from.

However, CLUSTERNODE does not work with a Veritas cluster, so the result is
that we get two parallel filespaces named  \\J2P-CLUSTER101\X$ and
\\J2P-CLUSTER101\X$ because X: sometimes resides on J2P-CLUSTER101
and some times on J2P-CLUSTER102.

From previous postings on this list I get the impression that there is
nothing
we can do to solve this, but I just want to be sure if this is true

How are other people who are running Veritas clusters and TSM solving this
?

This is a bad situation and one could fear (from a TSM person point of
view) that
it could pave the way for the introduction of Veritas backup software at
the site.

If CLUSTERNODE cannot easily be enhanced to support Veritas clusters as
well, all we would need would be a new statement in the option file,
perhaps called
MACHINENAME or so, where we could specify 'MACHINENAME WEB1' in the
optionfile and then have TSM use this name is the filespacing naming
process.

Regards

Peter



Re: Question about Veritas Cluster and filespace name

2002-12-11 Thread David Rigaudiere
Yes, i have exactly the same problem with Windows  Double-Take,
i would like change the way that TSM nam the the filespacing naming
process.


--
David  Rigaudiere  -+-  Administration  TSM + NT
Paris -+- 40, rue de Courcelles -+- 4e étage -+-
[EMAIL PROTECTED] -+- 01.5621.7802



|-+---
| |   Peter Bjoern|
| |   pebjn@WMDATASDC|
| |   .DK|
| |   Sent by: ADSM: |
| |   Dist Stor   |
| |   Manager|
| |   [EMAIL PROTECTED]|
| |   T.EDU  |
| |   |
| |   |
| |   11/12/2002 12:45|
| |   Please respond  |
| |   to ADSM: Dist  |
| |   Stor Manager   |
| |   |
|-+---
  
---|
  |
   |
  |To:  [EMAIL PROTECTED]   
   |
  |cc: 
   |
  |Subject: Question about Veritas Cluster and filespace name  
   |
  
---|




Hi

We have a situation with a Veritas cluster installed on two Win2000
servers.
The physical servers are called J2P-CLUSTER101 and J2P-CLUSTER102.
There is a virtual server defined with the name WEB1.

Each physical server has it's own C: drive which is backed up via a
schedule
that specifies 'C:' and 'systemobject' on the domain statement in the
option file.
The C: filespaces are given the names \\J2P-CLUSTER101\C$ and
\\J2P-CLUSTER102\C$ respectively.  This is all OK.

There is a shared drive X: which is associated with the WEB1 resource and
moves with WEB1 between the two physical servers so that it is available
either on J2P-CLUSTER101 or J2P-CLUSTER102.

X: is backed up from a virtual node defined in TSM called WEB1 and via a
separate TSM scheduler
service with an option file that  specifies X: on the domain statement.

If this had been a Microsoft Cluster installation, we would have specified
'CLUSTERNODE YES' in the optionfiles to make TSM name the X: filespace
with the virtual clustername (WEB1) so that it would have been named
\\WEB1\X$ regardless of which physical server it was backed up from.

However, CLUSTERNODE does not work with a Veritas cluster, so the result is
that we get two parallel filespaces named  \\J2P-CLUSTER101\X$ and
\\J2P-CLUSTER101\X$ because X: sometimes resides on J2P-CLUSTER101
and some times on J2P-CLUSTER102.

From previous postings on this list I get the impression that there is
nothing
we can do to solve this, but I just want to be sure if this is true

How are other people who are running Veritas clusters and TSM solving this
?

This is a bad situation and one could fear (from a TSM person point of
view) that
it could pave the way for the introduction of Veritas backup software at
the site.

If CLUSTERNODE cannot easily be enhanced to support Veritas clusters as
well, all we would need would be a new statement in the option file,
perhaps called
MACHINENAME or so, where we could specify 'MACHINENAME WEB1' in the
optionfile and then have TSM use this name is the filespacing naming
process.

Regards

Peter



---
This message (including any attachments) is confidential and may be
privileged. If you have received it by mistake please notify the sender by
return e-mail and delete this message from your system. Any unauthorised
use or dissemination of this message in whole or in part is strictly
prohibited. Please note that e-mails are susceptible to change.
ABN AMRO Bank N.V. (including its group companies) shall not be liable for
the improper or incomplete transmission of the information contained in
this communication nor for any delay in its receipt or damage to your
system. ABN AMRO Bank N.V. (or its group companies) does not guarantee that
the integrity of this communication has been maintained nor that this
communication is free of viruses, interceptions or interference.
---

Le présent message (y compris tous les éléments attachés) est confidentiel
et est destiné aux seules personnes qu'il vise. Si vous l'avez reçu par
erreur, merci de l'indiquer à son expéditeur par retour et de procéder à sa

Re: Question about Veritas Cluster and filespace name

2002-12-11 Thread Jim Kirkman
You know, there seems to be a similar issue with filespace naming with the new Netware 
cluster aware client. We're awaiting a server
upgrade to pursue it further. Seems as it the 'getservername' API call is at the heart 
of this.



David Rigaudiere wrote:

 Yes, i have exactly the same problem with Windows  Double-Take,
 i would like change the way that TSM nam the the filespacing naming
 process.

 --
 David  Rigaudiere  -+-  Administration  TSM + NT
 Paris -+- 40, rue de Courcelles -+- 4e étage -+-
 [EMAIL PROTECTED] -+- 01.5621.7802

 |-+---
 | |   Peter Bjoern|
 | |   pebjn@WMDATASDC|
 | |   .DK|
 | |   Sent by: ADSM: |
 | |   Dist Stor   |
 | |   Manager|
 | |   [EMAIL PROTECTED]|
 | |   T.EDU  |
 | |   |
 | |   |
 | |   11/12/2002 12:45|
 | |   Please respond  |
 | |   to ADSM: Dist  |
 | |   Stor Manager   |
 | |   |
 |-+---
   
|
   |  
 |
   |To:  [EMAIL PROTECTED] 
 |
   |cc:   
 |
   |Subject: Question about Veritas Cluster and filespace name
 |
   
|

 Hi

 We have a situation with a Veritas cluster installed on two Win2000
 servers.
 The physical servers are called J2P-CLUSTER101 and J2P-CLUSTER102.
 There is a virtual server defined with the name WEB1.

 Each physical server has it's own C: drive which is backed up via a
 schedule
 that specifies 'C:' and 'systemobject' on the domain statement in the
 option file.
 The C: filespaces are given the names \\J2P-CLUSTER101\C$ and
 \\J2P-CLUSTER102\C$ respectively.  This is all OK.

 There is a shared drive X: which is associated with the WEB1 resource and
 moves with WEB1 between the two physical servers so that it is available
 either on J2P-CLUSTER101 or J2P-CLUSTER102.

 X: is backed up from a virtual node defined in TSM called WEB1 and via a
 separate TSM scheduler
 service with an option file that  specifies X: on the domain statement.

 If this had been a Microsoft Cluster installation, we would have specified
 'CLUSTERNODE YES' in the optionfiles to make TSM name the X: filespace
 with the virtual clustername (WEB1) so that it would have been named
 \\WEB1\X$ regardless of which physical server it was backed up from.

 However, CLUSTERNODE does not work with a Veritas cluster, so the result is
 that we get two parallel filespaces named  \\J2P-CLUSTER101\X$ and
 \\J2P-CLUSTER101\X$ because X: sometimes resides on J2P-CLUSTER101
 and some times on J2P-CLUSTER102.

 From previous postings on this list I get the impression that there is
 nothing
 we can do to solve this, but I just want to be sure if this is true

 How are other people who are running Veritas clusters and TSM solving this
 ?

 This is a bad situation and one could fear (from a TSM person point of
 view) that
 it could pave the way for the introduction of Veritas backup software at
 the site.

 If CLUSTERNODE cannot easily be enhanced to support Veritas clusters as
 well, all we would need would be a new statement in the option file,
 perhaps called
 MACHINENAME or so, where we could specify 'MACHINENAME WEB1' in the
 optionfile and then have TSM use this name is the filespacing naming
 process.

 Regards

 Peter

 ---
 This message (including any attachments) is confidential and may be
 privileged. If you have received it by mistake please notify the sender by
 return e-mail and delete this message from your system. Any unauthorised
 use or dissemination of this message in whole or in part is strictly
 prohibited. Please note that e-mails are susceptible to change.
 ABN AMRO Bank N.V. (including its group companies) shall not be liable for
 the improper or incomplete transmission of the information contained in
 this communication nor for any delay in its receipt or damage to your
 system. ABN AMRO Bank N.V. (or its group companies) does not guarantee that
 the integrity of this communication has been maintained nor that this
 communication is free of viruses