We have configured DAG between our 2 x Exchange 2013 in an Active-Passive scenario giving us a DR site to failover to.
I have blocked activation on the DR site so that things dont just flip over when they want too.
We experienced an issue yesterday where our Exchange 2013 production server was rejecting mail stating that it was out of resources. This was a lie because the server had great RAM, CPU and disk utilisation. The only think I did notice though was in the EAC the copy queue lengths were very high and even now they are growing continuously.
When I do a Get-MailboxDatabaseCopyStatus I am seeing the copy queue length and reply queue length all 0.
When I do a Test-ReplicationHealth I see it fails on DatabaseRedundancy and DatabaseAvalibility. I know it is failing on the DatabaseRedundancy due to the copy queue length and it fails on the DatabaseAvalibility because I have blocked the activation in the DR site.
So I am confused why does EAC state the copy queue length as being very high but then a Test-MailboxDatabaseCopyStatus shows as 0 for both columns?
Also I have noticed that data is being replicated between the production and DR site but it is very slow! It is not utilizing the entire bandwidth and seems to be sending in sort bursts.
How does the replication of log shipping send in a DAG? Will the "Intersite replication SMTP Properties" in Active Directory Sites and Services have any bearing on this? Currently only 1 site is in the link and is set to replicate every 35 minutes..