Installing a local instance of SQL on a Cluster Node

Hi
I have an Active/Passive SQL 2000 Cluster - the OS is W3K Server.
I would like to install a local copy of SQL Server on the node that is
currently passive - if I start the SQL install process, it does give me
this option but I don't know what, if any, problems this might cause.
I appreciate if the currently active node fails, it will failover to
the node which also has a local version running as well.
Has anyone done this? I've tried searching groups and Microsoft without
success.
Thanks
Toby
It's OK to install local instances on cluster nodes. Just make sure the
instance name does not conflict with any instance on the cluster.
Tom
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
..
<tjbeaumont@.gmail.com> wrote in message
news:1160036955.164570.322690@.i3g2000cwc.googlegro ups.com...
Hi
I have an Active/Passive SQL 2000 Cluster - the OS is W3K Server.
I would like to install a local copy of SQL Server on the node that is
currently passive - if I start the SQL install process, it does give me
this option but I don't know what, if any, problems this might cause.
I appreciate if the currently active node fails, it will failover to
the node which also has a local version running as well.
Has anyone done this? I've tried searching groups and Microsoft without
success.
Thanks
Toby
|||<tjbeaumont@.gmail.com> wrote in message
news:1160036955.164570.322690@.i3g2000cwc.googlegro ups.com...
> Hi
> I have an Active/Passive SQL 2000 Cluster - the OS is W3K Server.
> I would like to install a local copy of SQL Server on the node that is
> currently passive - if I start the SQL install process, it does give me
> this option but I don't know what, if any, problems this might cause.
Yes, you can do it, but it is a violation of the HA spirit. Meaning, why in
the world would you want to screw around and do that when the company has
obviously felt that the other SQL instances (in the cluster) require HA?
Don't mess with the HA environment!
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp
|||That's true for production clusters. But we install standalone SQL instances
on a clustered node all the time on our test clusters.
Linchi
"Russ Kaufmann" wrote:

> <tjbeaumont@.gmail.com> wrote in message
> news:1160036955.164570.322690@.i3g2000cwc.googlegro ups.com...
> Yes, you can do it, but it is a violation of the HA spirit. Meaning, why in
> the world would you want to screw around and do that when the company has
> obviously felt that the other SQL instances (in the cluster) require HA?
> Don't mess with the HA environment!
>
> --
> Russ Kaufmann
> MVP - Windows Server - Clustering
> ClusterHelp.com, a Microsoft Certified Gold Partner
> Web http://www.clusterhelp.com
> Blog http://msmvps.com/clusterhelp
>
>
|||"Linchi Shea" <LinchiShea@.discussions.microsoft.com> wrote in message
news:6B70403B-35CF-4773-9160-8FFC9DDC8638@.microsoft.com...
> That's true for production clusters. But we install standalone SQL
> instances
> on a clustered node all the time on our test clusters.
So, your test environment doesn't mirror production?
Actually, I can see doing that.
Russ Kaufmann
MVP - Windows Server - Clustering
ClusterHelp.com, a Microsoft Certified Gold Partner
Web http://www.clusterhelp.com
Blog http://msmvps.com/clusterhelp

Installing A Failover Cluster

I have two matched servers, one will be Active and one will be Passive.
The servers have Windows 2000 Advanced Server and I have SQL Server 2000
Enterprise Edition.
My question is, do I also need a box called a Shared Disk Array, or can I
use another server I currently have that has lots of disk space on it?
Correction: I have Windows 2000 Advanced Server
"Vincel2k2" wrote:

> I have two matched servers, one will be Active and one will be Passive.
> The servers have Windows 2000 Advanced Server and I have SQL Server 2000
> Enterprise Edition.
> My question is, do I also need a box called a Shared Disk Array, or can I
> use another server I currently have that has lots of disk space on it?
>
|||You will indeed need shared storage.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://www.msmvps.com/clustering - Blog
"Vincel2k2" <Vincel2k2@.discussions.microsoft.com> wrote in message
news:FF446D59-A73C-43EA-9EC2-4B9D03CFD51B@.microsoft.com...[vbcol=seagreen]
> Correction: I have Windows 2000 Advanced Server
> "Vincel2k2" wrote:
|||Hi
You need to get your self a Storage Area Network (SAN) to store your data
on.
For clustering, make sure that your hardware is on the Windows Hardware
Compatibility List (HCL)
Read the documentation on
http://www.microsoft.com/windowsserv...g/default.mspx
to get an understanding.
Regards
Mike Epprecht, Microsoft SQL Server MVP
Zurich, Switzerland
IM: mike@.epprecht.net
MVP Program: http://www.microsoft.com/mvp
Blog: http://www.msmvps.com/epprecht/
"Vincel2k2" <Vincel2k2@.discussions.microsoft.com> wrote in message
news:FF446D59-A73C-43EA-9EC2-4B9D03CFD51B@.microsoft.com...[vbcol=seagreen]
> Correction: I have Windows 2000 Advanced Server
> "Vincel2k2" wrote:
I[vbcol=seagreen]
|||I also have two matched servers with Win2003EntSrv that I would like to
install an active/passive failover cluster of Yukon on. Since this is a lab
environment and I don't have the $$ for a SAN, will it work with a Network
Storage HD? I realize the performance won't be great, but I'm looking for
the experience in how to build/configure the environment.
Any tips?
An alternative question for the lab environment is ... how would you do this
in two VPC? Or would you steer clear?
Thanks
Al
"Mike Epprecht (SQL MVP)" wrote:

> Hi
> You need to get your self a Storage Area Network (SAN) to store your data
> on.
> For clustering, make sure that your hardware is on the Windows Hardware
> Compatibility List (HCL)
> Read the documentation on
> http://www.microsoft.com/windowsserv...g/default.mspx
> to get an understanding.
> Regards
> --
> Mike Epprecht, Microsoft SQL Server MVP
> Zurich, Switzerland
> IM: mike@.epprecht.net
> MVP Program: http://www.microsoft.com/mvp
> Blog: http://www.msmvps.com/epprecht/
> "Vincel2k2" <Vincel2k2@.discussions.microsoft.com> wrote in message
> news:FF446D59-A73C-43EA-9EC2-4B9D03CFD51B@.microsoft.com...
> I
>
>
|||Virtual Server 2005 will work nicely for this test. See
http://www.hands-on-labs.com/microsoft.asp?ID=4477 for the steps
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"al_stark_born" <al_stark_born@.discussions.microsoft.com> wrote in message
news:E579B181-45E6-4DF2-8BCA-E7F1D2DD44C7@.microsoft.com...[vbcol=seagreen]
>I also have two matched servers with Win2003EntSrv that I would like to
> install an active/passive failover cluster of Yukon on. Since this is a
> lab
> environment and I don't have the $$ for a SAN, will it work with a Network
> Storage HD? I realize the performance won't be great, but I'm looking for
> the experience in how to build/configure the environment.
> Any tips?
> An alternative question for the lab environment is ... how would you do
> this
> in two VPC? Or would you steer clear?
> Thanks
> Al
> "Mike Epprecht (SQL MVP)" wrote:
sql

Installing a cluster on an existing instance of SQL 2000 SP3A

I have been "tapped" to install a passive node on an instance of SQL
that is not currently configured for clustering on a W2003. Can I
simply install clustering services on the current production instance,
then install a new instance on what would be the passive node? Then add
it via cluster admin?
Can I run cluster communications, heartbeat, etc on the public NICs..I
do not have two interfaces on each server.
Any information would be greatly appreciated as I have only done new
deployments and this being a prod server I do not want to try the trial
and error approach.
Thanks,
Hulicat
You can use the current active node as a cluster member. You can then
install a clustered instance of SQL and migrate the databases. The final
step is to turn off the original instance. Possible, Yes. Recommended, NO.
This entire project sounds like a low-availability solution. You are
reusing equipment instead of purchasing a certified cluster configuration.
The one-NIC cluster is possible but falls into the "worst practices"
category. Not having a test platform, even a virtual one, is asking for
even more trouble.
Clustering is a hardware and systems tool to help achieve increased system
availability. Shortcutting the hardware and the procedures will result in
lower availability rather than higher availability.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"Hulicat" <dennis_A_white@.yahoo.com> wrote in message
news:1162242129.910915.61910@.e3g2000cwe.googlegrou ps.com...
>I have been "tapped" to install a passive node on an instance of SQL
> that is not currently configured for clustering on a W2003. Can I
> simply install clustering services on the current production instance,
> then install a new instance on what would be the passive node? Then add
> it via cluster admin?
> Can I run cluster communications, heartbeat, etc on the public NICs..I
> do not have two interfaces on each server.
> Any information would be greatly appreciated as I have only done new
> deployments and this being a prod server I do not want to try the trial
> and error approach.
> Thanks,
> Hulicat
>
|||Thanks for that!
I am going tp lobby for new hardware or scratch the idea.
It's a small SPS portal.
Regards~~~
Geoff N. Hiten wrote:[vbcol=seagreen]
> You can use the current active node as a cluster member. You can then
> install a clustered instance of SQL and migrate the databases. The final
> step is to turn off the original instance. Possible, Yes. Recommended, NO.
> This entire project sounds like a low-availability solution. You are
> reusing equipment instead of purchasing a certified cluster configuration.
> The one-NIC cluster is possible but falls into the "worst practices"
> category. Not having a test platform, even a virtual one, is asking for
> even more trouble.
> Clustering is a hardware and systems tool to help achieve increased system
> availability. Shortcutting the hardware and the procedures will result in
> lower availability rather than higher availability.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
>
> "Hulicat" <dennis_A_white@.yahoo.com> wrote in message
> news:1162242129.910915.61910@.e3g2000cwe.googlegrou ps.com...