Hi all,

I have 2 Exchange 2k13 nodes in DAG, and I noticed that I have 2 DB acive on node1 and only passive DB on node2.

If node1 goes down, it will be able to do a failover of both the mailbox DBs to node 2? Or it's mandatory that each node has 1 DB active and the other DB as passive, so if one node goes down, the failover has to mount as active only one DB?

In this case, it is ok to launch this script to balance the DBs?

> C:\Program Files\Microsoft\Exchange Server\V15\Scripts>.\RedistributeActiveDatabases.ps1 -DagName <<name>> -BalanceDbsBy ActivationPreference

Thank you for your support

Hi all,

I have 2 Exchange 2k13 nodes in DAG, and I noticed that I have 2 DB acive on node1 and only passive DB on node2.

If node1 goes down, it will be able to do a failover of both the mailbox DBs to node 2? Or it's mandatory that each node has 1 DB active and the other DB as passive, so if one node goes down, the failover has to mount as active only one DB?

In this case, it is ok to launch this script to balance the DBs?

> C:\Program Files\Microsoft\Exchange Server\V15\Scripts>.\RedistributeActiveDatabases.ps1 -DagName <<name>> -BalanceDbsBy ActivationPreference

Or there are other command/procedure that are better than this one?

Thank you for your support

That depends on a few things.  When you built your Exchange Servers did you size them so that they can handle the load?  Are they in the same site?  

it really doesn't matter which node the databases run on to be honest as long as you are sized properly.  Also, if you only have 2 databases, it's probably easier to move the databases in the EAC (like 1 click) or the EMS using Move-ActiveMailboxDatabase 

https://technet.microsoft.com/en-us/library/Dd298068(v=EXCHG.150).aspx

Hi,

yes, the nodes are in the same site and are sized to handle the load in case of failover.

My question was born because it happened that while I was patching a node, I was expecting that it will automatically failover to the other node...but it seemed to not success in this operation, resulting in Exchange access down for 10-15 minutes!

So I'm trying to serch the causes of this inconvenient...but, if you say that thte balancing of the DBs could not be the cause, I have to watch to something else, maybe our network load balancer...

Hi,

yes, the nodes are in the same site and are sized to handle the load in case of failover.

My question was born because it happened that while I was patching a node, I was expecting that it will automatically failover to the other node...but it seemed to not success in this operation, resulting in Exchange access down for 10-15 minutes!

So I'm trying to serch the causes of this inconvenient...but, if you say that thte balancing of the DBs could not be the cause, I have to watch to something else, maybe our network load balancer...

Hi Luca,

Thank you for your question.

Issue1:

Because you have 2 node in DAG, it will has a FSW in your organization if we want to DAG failover.

Issue 2:

We could run the script which you refer to. In addition, we suggest you distribute active DB on different node for mailbox performance.

If there are any questions regarding this issue, please be free to let me know.

Best Regard,

Jim

Hi,

We need to know, few things first. Is it a multirole or you have separate CAS servers.?

As it is a 2 node cluster, there will be a FileShareWitness Server automatically created on the CAS most likely, if you havn't specified anything else.

Run this cmdlet to identify the current FSW

Get-ClusterResource

So , if one node is down, the FSW should be available for node 2 to successfully failover and continue the service. If FSW goes down as well, then you will loose quorum and DBs will dismount.

Read this for more understanding and steps:

How does Dynamic Quorum work for a two Node DAG