SCOM Deployment DR

Question

If I want to deploy SCOM for HA and DR. Can I put OM DB and DWH DB in a 2-node SQL Always On cluster with mirroring to the DR site? Or I can set both OM DB and DWH DB instance on a 2-note cross sites cluster - active-active? I will only monitor Windows agents and there will not be any need for gateway servers. I am planning to put 2 MS in primary site and 1 MS in DR site as a management group? The Windows agent will automatically report to any MS in the management group should any MS fail right? I read that MS in the DR site should be set to passive. Is there any need to do this? For the web console server component, operation console server component and reporting server component, can I deploy them in the SQL cluster? Thanks.

Answer 1

Hi @ChongSunLimAP-4498,

Can I put OM DB and DWH DB in a 2-node SQL Always On cluster with mirroring to the DR site? Or I can set both OM DB and DWH DB instance on a 2-note cross sites cluster - active-active?

Yes, the Operations Manager database and data warehouse supports these SQL Server Always On scenarios, you'll find more information here:

SQL Server Design Considerations - SQL Server Always On
https://learn.microsoft.com/en-us/system-center/scom/plan-sqlserver-design?view=sc-om-2019#sql-server-always-on

I am planning to put 2 MS in primary site and 1 MS in DR site as a management group? The Windows agent will automatically report to any MS in the management group should any MS fail right?

Yes this should work without any issues, Windows agents failover automatically to any management server.

I read that MS in the DR site should be set to passive. Is there any need to do this?

It doesn't have to be passive, this is up to you how it will end up, you can control this if you want.

For the web console server component, operation console server component and reporting server component, can I deploy them in the SQL cluster?

I would strongly recommend not to do this, the Reporting Server cannot be highly available, so you can for example install it on a standalone server, or on one of the SQL Server cluster nodes. The Web Console and Operations Console can be installed on any server which meets the system requirements, but I would not recommend installing it on the SQL Server cluster.

Some useful documentations:

High Availability and Disaster Recovery
https://learn.microsoft.com/en-us/system-center/scom/plan-hadr-design?view=sc-om-2019

Planning a Management Group Design
https://learn.microsoft.com/en-us/system-center/scom/plan-mgmt-group-design?view=sc-om-2019

----------

(If the reply was helpful please don't forget to upvote or accept as answer, thank you)

Best regards,
Leon

Answer 2

Hi @ChongSunLimAP-4498,

here are my 2 cents in addition to the thorough reply of Leon: usually if you have a DR site or a second site, it is recommended to prevent your agents from failing over to it. Usually this has lots of reasons, like firewall ports and connectivity in general. You should also remove your DR management server from any resource pools.
All this is clearly described here:

High Availability and Disaster Recovery
https://learn.microsoft.com/en-us/system-center/scom/plan-hadr-design?view=sc-om-2019

Here are some of the important pieces of information:

If this approach isn't acceptable, you can deploy management servers in your secondary data center for on-standby recovery. Remove them as members of the three primary resources pools - All Management Servers Resource Pool, Notifications, and AD Assignment. This also includes any custom resource pool, which may include management servers hosted in the primary data center and need to continue to function as part of the recovery plan. The System Center Data Access, System Center Configuration Management, and Microsoft Monitoring Agent services should be stopped and set to manual or disable and only started in a disaster recovery scenario.

And here is also the exact reference to the other recommendation:

If one site goes offline, the agent will fail over to the management server in another site, assuming that the agent’s failover configuration allows this. Reconfigure the Windows agents to cache only management servers in your primary data center that should manage them to prevent them from attempting to failover to a management server in the secondary data center, which would only delay recovery and reporting.

So, I would recommend you to configure your Agents to fail over only to management servers within your Primary site. This can very easily be accomplished with PowerShell:

Assigning Gateways and Agents to Management Servers using PowerShell
https://kevinholman.com/2018/08/06/assigning-gateways-and-agents-to-management-servers-using-powershell/

Hope I can be of help.

----------

(If the reply was helpful please don't forget to upvote or accept as answer, thank you)
Regards,
Stoyan