![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
How do I add a new instance to an existing 2 node cluster? I have two Server 2008 nodes supporting 2 clusters (active/active); cluster1 running as the default instance & cluster2\inst1. Now I want to create cluster2\inst2 using the same shared disk and resource group as cluster2\inst1. How do I do that without creating cluster3\inst2? |
#3
| |||
| |||
|
|
You can't use a drive that is already being used by a different clustered instance. For the new instance, you need at least one new drive in its own resource group. Linchi "BMarxhausen" wrote: How do I add a new instance to an existing 2 node cluster? I have two Server 2008 nodes supporting 2 clusters (active/active); cluster1 running as the default instance & cluster2\inst1. Now I want to create cluster2\inst2 using the same shared disk and resource group as cluster2\inst1. How do I do that without creating cluster3\inst2? |
#4
| |||
| |||
|
|
Why not? I have a staging and development instance that I want to share the same drives and network name. If one fails over the other one can/should too. "Linchi Shea" wrote: You can't use a drive that is already being used by a different clustered instance. For the new instance, you need at least one new drive in its own resource group. Linchi "BMarxhausen" wrote: How do I add a new instance to an existing 2 node cluster? I have two Server 2008 nodes supporting 2 clusters (active/active); cluster1 running as the default instance & cluster2\inst1. Now I want to create cluster2\inst2 using the same shared disk and resource group as cluster2\inst1. How do I do that without creating cluster3\inst2? |
#5
| |||
| |||
|
|
Sure, conceptually why not if you indeed want them to go together. But there is only so much flexibility one can design into a product and still be profitable. I guess the restriction is by design, and for good enough reasons. After all, this is an HA feature, and as such you'd rather it be clean. Then, you may want to ask yourself, do you really need to have them fail over together in a DEV/UAT environment? If it does work that way, does it really cause any real problem for you? I don't think so. Linchi "BMarxhausen" wrote: Why not? I have a staging and development instance that I want to share the same drives and network name. If one fails over the other one can/should too. "Linchi Shea" wrote: You can't use a drive that is already being used by a different clustered instance. For the new instance, you need at least one new drive in its own resource group. Linchi "BMarxhausen" wrote: How do I add a new instance to an existing 2 node cluster? I have two Server 2008 nodes supporting 2 clusters (active/active); cluster1 running as the default instance & cluster2\inst1. Now I want to create cluster2\inst2 using the same shared disk and resource group as cluster2\inst1. How do I do that without creating cluster3\inst2? |
#6
| |||
| |||
|
|
Main issue I have is a lack of drive letters. I had intentions of installing more instances, but run out of letters quick if I seperate data, logs, backup, tempdb for each instance, and already allocate a drive for DTC, Quorum, sysvol, and DVD. Not to mention drive mappings that fire via login. Seems to me an artificial restriction. If the instances can share drives via directory seperation, why force them to have their own resource group with dedicated drives? Maybe a licensing issue? Are 2 licenses required for cluster1\inst1 + cluster2\inst2 vs. cluster1\inst1 + cluster1\inst2? Still this could be handled with the requirement to have seperate network name per cluster yet still allow the drives to be shared. "Linchi Shea" wrote: Sure, conceptually why not if you indeed want them to go together. But there is only so much flexibility one can design into a product and still be profitable. I guess the restriction is by design, and for good enough reasons. After all, this is an HA feature, and as such you'd rather it be clean. Then, you may want to ask yourself, do you really need to have them fail over together in a DEV/UAT environment? If it does work that way, does it really cause any real problem for you? I don't think so. Linchi "BMarxhausen" wrote: Why not? I have a staging and development instance that I want to share the same drives and network name. If one fails over the other one can/should too. "Linchi Shea" wrote: You can't use a drive that is already being used by a different clustered instance. For the new instance, you need at least one new drive in its own resource group. Linchi "BMarxhausen" wrote: How do I add a new instance to an existing 2 node cluster? I have two Server 2008 nodes supporting 2 clusters (active/active); cluster1 running as the default instance & cluster2\inst1. Now I want to create cluster2\inst2 using the same shared disk and resource group as cluster2\inst1. How do I do that without creating cluster3\inst2? |
![]() |
| Thread Tools | |
| Display Modes | |
| |