We run 6 virtualized Exchange 2019 Mailbox servers within a single DAG. Our configuration ensures that each database has two copies. The arrangement is such that every pair of servers share the same database copies. This means:
- Server 1 shares database copies with Server 2
- Server 3 shares with Server 4
- Server 5 shares with Server 6
Our storage team has informed us of an upcoming maintenance session where the storage connected to all these Exchange servers will be temporarily offline. I want to ensure a clean shutdown of all the Exchange Mailbox servers prior to the outage, and safely bring them online post-maintenance, with an added check to verify the storage's integrity.
Proposed Shutdown and Startup Sequence:
Before Storage Maintenance:
- Place Exchange Servers 2, 4, and 6 into Exchange maintenance mode, and then shut them down. This will keep all DB copies active on servers 1, 3, and 5.
- Dismount all databases.
- Stop all Exchange services on servers 1, 3, and 5.
After Storage Maintenance:
- First, inspect and verify the storage's health.
- If everything checks out, initiate all Exchange services on servers 1, 3, and 5, and verify their status.
- Mount all databases.
- Power on Exchange servers 2, 4, and 6.
- Remove Exchange Servers 2, 4, and 6 from Exchange maintenance mode.
Feedback Request:
I would greatly appreciate insights or recommendations to enhance this approach. Are there any precautions or additional steps I should consider?
