none
Exchange 2016 DAG Maintenance Problems RRS feed

  • Question

  • Hi,

    i got a question about DAG Maintenance.

    I got 2 Exchange 2016 CU13 Enterprise and on qourom, no DAC Mode.

    Outlook Online, not cached.

    Both Servers has 65 GB RAM, 1TB System, a 12 TB and a 6 TB DB Partition ntfs (Not ReFS) and 1 TB Log.

    (10RPM SAS), and bot Servers has 2 Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz with 10 Cores per device!

    36 DBs 18 are active per Server, not bigger as 500 GB. All DBs are round about 10TB

    ~1200 DBs

    Our consultans said, our Server were to big for our environment...

    But when I try to set on in maintenance to update him, i got massiv outlook-connection and performance problems.

    Microsoft says, we have just to use: 

    Move-ActiveMailboxDatabase -Server MBX1

    and fine. This works?

    We do this:

    Set-ServerComponentState MBX1 -Component HubTransport -State Draining -Requester Maintenance
    Set-ServerComponentState MBX1 -Component UMCallRouter -State Draining -Requester Maintenance 
    Restart-Service MSExchangeTransport 
    Redirect-Message -Server MBX1 -Target MBX2.hamburg.hih.local confirm:$false
    Suspend-ClusterNode MBX1
    get-clusternode MBX1
    Set-MailboxServer MBX1 -DatabaseCopyActivationDisabledAndMoveNow $True 
    Set-MailboxServer MBX1 -DatabaseCopyAutoActivationPolicy Blocked 
    Set-ServerComponentState MBX1 -Component ServerWideOffline -State Inactive -Requester Maintenance
    Restart-Service MSExchangeTransport  

    The thing is.. when I do server failover during workhours, it almost collapsed CPU Red at 99%/100% normally he calms down, after 30 Minutes.. 

    At night it is ok, when no Backup runs. Sometime we try to set the Server in Maintenance at night and update at the Morning, but now (we grow a little)an this doesnt help, yet.

    Is it a hardware bottleneck?  

    Wednesday, August 14, 2019 10:18 AM

All replies

  • I forget, Loadbalanced with Kemp and many active Sync-Devices..
    Wednesday, August 14, 2019 10:50 AM
  • No, definitely not a hardware bottleneck, from what I can see. What antivirus do you have on the server?

    You should be able to get a good idea of what is using up the CPU cycles in task manager's resource monitor and the list of processes. Once you know which component is using up the CPU, you'll be able to more easily track down the cause. Performance issues like this are notoriously difficult to troubleshoot.

    Wednesday, August 14, 2019 1:36 PM
  • Hmm we got Trend Micro and Trend Micro ScanMail for Exchange, but I think the got the right scan exceptions.
    Wednesday, August 14, 2019 4:23 PM
  • It is recommended to use the script StartDagServerMaintenance.ps1 to perform any type of software or hardware maintenance, it assists with moving all active databases off the server. It also moves all critical DAG support functionality, such as the Primary Active Manager (PAM) role, and blocks them from moving back to the server before maintenance is complete.

    For the details steps,see Performing maintenance on DAG members.

    Regards,

    Manu Meng


    Please remember to mark the replies as answers if they helped. If you have feedback for TechNet Subscriber Support, contact tnsf@microsoft.com.

    Thursday, August 15, 2019 6:56 AM
    Moderator
  • Just checking in to see if above information was helpful. Please let us know if you would like further assistance.

    Regards, 

    Manu Meng


    Please remember to mark the replies as answers if they helped. If you have feedback for TechNet Subscriber Support, contact tnsf@microsoft.com.

    Monday, August 19, 2019 9:37 AM
    Moderator