Replication question

[ashman70]

I've been suffering from a bug affecting Esxi 5.1 patch 2 specifically related to intel E1000 network cards, since I am unable to reboot the host until the weekend the workaround has been to replace the network card in the vm with a VMXNET2 adapter. Once I do this, if I do an incremental replication, will the replica update this adapter or not?

Also, I had to failover to a replica this morning, however when I did it, the original VM did not power off, the replica just powered on, is this normal behaviour? I powered the original VM off manually. I waited about 2-3 minutes for the original VM to power off on its own but it didn't and it was a bit of timing sensitive issue so I had to power off the original VM so users could connect to the replica.

[foggy]

I've been suffering from a bug affecting Esxi 5.1 patch 2 specifically related to intel E1000 network cards, since I am unable to reboot the host until the weekend the workaround has been to replace the network card in the vm with a VMXNET2 adapter. Once I do this, if I do an incremental replication, will the replica update this adapter or not?

You should be able to continue replication after editing replica VM configuration file.

Also, I had to failover to a replica this morning, however when I did it, the original VM did not power off, the replica just powered on, is this normal behaviour? I powered the original VM off manually. I waited about 2-3 minutes for the original VM to power off on its own but it didn't and it was a bit of timing sensitive issue so I had to power off the original VM so users could connect to the replica.

Yes, this is normal behavior, original VM is not automatically powered off during failover. Typically, in a DR situation, it is already down (otherwise, why to failover?).

[veremin]

Also, I had to failover to a replica this morning, however when I did it, the original VM did not power off, the replica just powered on, is this normal behaviour? I powered the original VM off manually. I waited about 2-3 minutes for the original VM to power off on its own but it didn't and it was a bit of timing sensitive issue so I had to power off the original VM so users could connect to the replica.

In order to achieve what you're after you should have used Planned Failover, instead. Thanks.

[ashman70]

So I have made configuration changes to the original vm, in this case I have changed the network adapter from E1000 to VMXNET2, however the replica is still on the E1000. If I fail back to production and commit the fail back, then I'll wind up with the E1000 again, correct? When I changed the network card on the terminal server it must of somehow unbound the terminal server service since it was originally bound to the E1000 network card, changing the network card somehow killed that and no one could connect, however they can connect to the replica just fine. I just want to make sure when I fail back to production, everything should work as its working now.

[ashman70]

Also I need to patch two Esxi hosts, one of them is currently running this replica, what is the process so I can reboot the host? Do I just power off the replica vm, then reboot the host then power the vm back up? What are the ramifications?

[foggy]

I'm not sure failback will work either, after changing the production VM NIC.

[ashman70]

Is there not a process that will fail back creating a new vm from the replica?

[foggy]

I meant failback to original location. Alternative location should work.

[ashman70]

I don't have an alternate location, there are only two hosts, the production and the backup, the replica is currently running off the backup host.

[foggy]

By alternative location I mean a new VM, not the original one (it may reside on the same host as well).

[ashman70]

Ah ok, then that is what I may have to do then.