I am trying to do the following and want to make sure I am following the correct steps.
Critical servers that need to stay up and running so replication jobs are running for them. I need to patch the OS on these servers. Is the best way to do this is start up the replicas to a different network on my DR site and then patch the VMs on the PROD site. Once patching is completed undo the failover to PROD site?
If there is another way to do this please let me know as well.
Thanks
-
- Novice
- Posts: 8
- Liked: never
- Joined: Jun 15, 2016 7:33 pm
- Full Name: Ray McDonald
- Contact:
-
- Veeam Software
- Posts: 2853
- Liked: 654 times
- Joined: Jun 28, 2016 12:12 pm
- Contact:
Re: Veeam Replication/Guest VMs OS patching
Hi rmcdonald75,
Potentially yes, however please pay close attention here as Undo Failover does NOT copy changes from DR to Production; changes made during the failover will be lost with Undo.
If the machines don't accumulate any data that needs to be replicated back, then your plan can work just fine; VBR will warn about this before the Undo Failover takes place (as you can see in the screenshot on the linked User Guide page)
If you need to retain the changes made to the replica while it was active, Failback is how you accomplish this, but note that Failback will have unavoidable downtime as noted in the workflow. For such cases, a maintenance window for the machines is typically the best approach.
Potentially yes, however please pay close attention here as Undo Failover does NOT copy changes from DR to Production; changes made during the failover will be lost with Undo.
If the machines don't accumulate any data that needs to be replicated back, then your plan can work just fine; VBR will warn about this before the Undo Failover takes place (as you can see in the screenshot on the linked User Guide page)
If you need to retain the changes made to the replica while it was active, Failback is how you accomplish this, but note that Failback will have unavoidable downtime as noted in the workflow. For such cases, a maintenance window for the machines is typically the best approach.
David Domask | Product Management: Principal Analyst
Who is online
Users browsing this forum: Baidu [Spider] and 18 guests