May 03 2019 06:59 AM
Hello
we are planning for shutdown all server in DC 01 to maintain electronic .
Current environment exchange server have
- Active directory : only one site in both DC 01 and DC 02 (can call DR) and both site has active directory
- 6 server mailbox (CAS+MBX) in DC and 3 server mailbox (CAS+MBX) in DR . all physical
- all mailbox database has 1 copy in 3 server in DR
- all server mailbox the same subnet in both site and the same site on active directory
- one CAS server (for ECP admin) is VM current in DC and can migrate to DR manual
please help me the procedure how can shutdown all 6 server that DAG can normal and all database mount in DR
Best Regards,
Thanks
May 03 2019 09:16 AM
Hi @Tien Ngo Thanh,
The best practice is that you have 2 Active Directory Sites and in each AD Site you need to have at least a Global Catalog. If you have that you can shutdown on Datacenter and you you have Exchange with replica in place it will move that. You can also see the DAC mode on Exchange to see if is correct configuration within this scenario.
When you startup an Exchange environment you will need at least on Global Catalog up in the AD site that the exchange nodes resides.
May 03 2019 09:42 AM
Hi
all active directory has global catalog enable .but current five role FSMO in DC , when shutdown i think i will change all FSMO to DR site
DAG configure
DatacenterActivationMode : DagOnly
AllowCrossSiteRpcClientAccess : False
WitnessServer : cas01.test.com
WitnessDirectory : E:\DagWitnessDirectory
Best Regards,
Thanks
May 03 2019 10:00 AM
hello tien. what version of exchange are you running? from what it sounds like, you will want to run something like stop-databaseavailabilitygroup DAG01 -activedirectorysite "DC01", which will put the DC01 site servers in the stopped servers list when you run "get-databaseavailabilitygroup DAG01". Then you run a restore-databaseavailabilitygroup DAG01 to reform the DAG with only the DC02 site servers in the dag. for more information, search for "exchange datacenter switchover"
May 03 2019 10:05 AM
Current active directory is windows 2012 r2 and exchange is 2013 CU20
May 03 2019 10:06 AM
Hi @Tien Ngo Thanh,
About DAC, please read this https://docs.microsoft.com/en-us/exchange/high-availability/database-availability-groups/dac-mode?vi...
AllowCrossSiteRpcClientAccess, must be enabled.
About Witness you must have one per datacenter.
Best Regards,
May 03 2019 10:16 AM - edited May 03 2019 10:19 AM
@NunoAriasSilva : with current exchange DAG both site the same subnet and the same site in Active Directory then should enable AllowCrossSiteRpcClientAccess ?
about Witness so i must install one more cas in DR and set alterwitness to CAS in DR ?
actual all server exchange the same site only difference about physical location because the same subnet .
May 03 2019 10:20 AM
Hi @Tien Ngo Thanh,
As my first post, the best practice is to have separate AD sites and IP Addresses and have Global Catalogs on each site and have one mote Witness on DR.
Best regards
May 05 2019 03:16 AM
@NunoAriasSilva : but current my exchange in tow data center use the same subnet , and another server (not exchnage also use the same subnet as sharepoint...) should we can not separate AD sites , and backbone between two high bandwidth . please recommend help me ?
My DAG in DAC Mode enable and DAG the same subnet , the same site , only difference location
May 06 2019 02:35 PM
In short, you need to remove all the dependencies before shutting down servers.
- Removing HA (DAG), move all mailboxes to server that will be active
- Move all Frontend settings to server that will be active
- Make sure those servers are not meant to send out emails in sender connector
- No DNS records present for those servers, like webmail, mail, autodiscover.
Regards,
Akshay
Exchange Admin - Apps4Rent
May 07 2019 04:21 PM
Hello
I just list some document and i think do that these step below
the first time remove member server in F5 load balancing after that
1. Check
Get-MailboxDatabaseCopyStatus -Server MBXV01
2. Active All Database to site dr
3. Run Get-ClusterGroup and Move-ClusterGroup –cluster DAG –name “Cluster Group” –node MBXServerDR
4. After that you could run to check
Get-DatabaseAvailabilityGroup -Status -Identity DAG | fl name,primaryActiveManager
5. Then, run Stop-DatabaseAvailabilityGroup -Identity DAG -MailboxServe MBX01 (for all mailbox in DC)
6. Power off each of Exchange server in DC
Thanks