• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 555
  • Last Modified:

4 Node Exchange Cluster

I need to know based on what criteria, the exchange server cluster resources are distributed among the 3 Active nodes.

in 2 nodes active/passive , one node has Zero (0) resources and the other has all resources
in 4 nodes 3 Active/1 passive, it is not easy to tell which specific resources  should be assigned to each active node.

Let's say Exchange cluster uses 20 resources total, how are those 20 resources assigned to each active node?

Thanks
0
jskfan
Asked:
jskfan
  • 5
  • 3
1 Solution
 
jskfanAuthor Commented:
It s exchange 2003
0
 
jskfanAuthor Commented:
I am talking about the Cluster that is applicable to exchange 2003 similar to SQL cluster.
I know that in Exchange 2007/2010 the concept of clustering is completely different than in Exchange 2003.

0
Problems using Powershell and Active Directory?

Managing Active Directory does not always have to be complicated.  If you are spending more time trying instead of doing, then it's time to look at something else. For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why

 
jskfanAuthor Commented:
any updates???
0
 
NenadicCommented:
Hi jskfan, the clustering resources are grouped together into a cluster group. That group as a whole gets assigned to a node. For example, let's say you have those four nodes and you want to run three Exchange clusters. You would then have a minimum of:
- 3 x Exchange System Attendant
- 3 x Network name
- 3 x IP address
- 3 x disk resources (there are likely to be more of these).
On top of this, you would have a default Windows cluster group with a network name and an IP address. These resources would exist in a total of four groups - most likely:
- Node 1: Windows default cluster group
- Node 1: Exchange Cluster 1
- Node 2: Exchange Cluster 2
- Node 3: Exchange Cluster 3
- Node 4: passive.
If all four Nodes are considered as hosts for all four clusters and you then lose Node 1, both Windows and Exchange Cluster 1 would fail over to Node 4. If you were to then lose Node 2, it is likely that Exchange Cluster 2 would fail over to Node 3. You are then effectively running two Exchange Servers off the same node. If that is a likely scenario, you may want to beef up your nodes.

As Nodes come back, your fail back preferences will determine whether cluster groups go back to their original hosts.

Hope that helps.
Good luck.
0
 
jskfanAuthor Commented:
Nenadic: Good Explanation!!!

where I need to  clear any ambiguity, is when you said:
<<you would have  the default Windows cluster group with a network name and an IP address. >>

When all 4 nodes are added to the cluster, are we supposed to manually add the network name and an IP address on each node even to th epassive node??

THEN

Go to 3 out of the 4 nodes and add :
- 3 x Exchange System Attendant
- 3 x Network name
- 3 x IP address
- 3 x disk resources
 ???



0
 
NenadicCommented:
Windows cluster group is your starting point, irrespective of Exchange. It gets created when you start Windows Clustering. So, the name and IP address wouldn't be used by Exchange users. I'm not sure if this is the explanation you are looking for.

So, a 4-node cluster could run in a Active/Active/Active/Passive setup, where one node is available to pick up a cluster group if an active node fails. In other words, that would be a single server resilience.
Alternatively, it could run as Active/Active/Passive/Passive, where two nodes are available to pick up the load. So, you could lose two full servers before any drop in performance is experienced.

Looking at first option, you'd create:
1. Windows Group with network name, IP address etc. - can run on all four nodes. Preferred node A, for example.
2. Exchange Group 1 with Exchange network name 1, Exchange IP address 1, disk resources 1a, 1b, 1c, 1d and Exchange SA 1. Preferred node A.
3. Exchange Group 1 with Exchange network name 2, Exchange IP address 2, disk resources 2a, 2b, 2c, 2d and Exchange SA 2. Preferred node B.
4. Exchange Group 1 with Exchange network name 3, Exchange IP address 3, disk resources 3a, 3b, 3c, 3d and Exchange SA 3. Preferred node C.

In other words, you add resources to cluster groups and you specify where groups can be activated and, optionally, which nodes should be preferred (in other words, if they fail and come back - the cluster group will automatically fail back to them).
0
 
jskfanAuthor Commented:
Excellent Explanation!!!
0
 
NenadicCommented:
Thanks, jskfan.
0

Featured Post

How to Use the Help Bell

Need to boost the visibility of your question for solutions? Use the Experts Exchange Help Bell to confirm priority levels and contact subject-matter experts for question attention.  Check out this how-to article for more information.

  • 5
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now