Nimble and VMware VVol replication

I touched VVol replication on 3PAR in an earlier post. The scripts need some minor modification to work with Nimble arrays.

But first, remember when running these scripts not to have an existing session to your vCenter, or your will have some things happening twice when running the scripts.

With PowerCLI 11 you no longer need the SourceVvolIdMap to bring up the VMs on site 2, so that part you can remove from your scripts.

The Nimble documentation states that the replication when using the vCenter Nimble plugin and vCenter SPBM replication are mutually exclusive. The documentation doesn’t go into much detail about the differences, and they are configured almost the same.
The red square below, if you omit that one you can only use vCenter Nimble plugin.
If you add it, you can only use SPBM replication.Capture.PNG

If you go with SPBM replication you also have to set replication groups on VMs, just like with 3PAR and others using SPBM replication. This option is not visible if you go for the plugin.

Another difference with Nimble compared to 3PAR is that the target SPBM group will have the same name as the source. So in the scripts I had to to add an extra check when stopping a TestFailOver to only hit the target group and not source as well. About line 103 in DR-StopTestFailover.ps1
$inTestGroup = $allOtherGroups | where-object { $ -eq $groupId -and $_.State -eq “InTest”}

Can the vCenter Nimble plugin be used for Disaster Recovery? The documentation clearly states that it is not intended for it.
In what areas is it lacking for DR? In my opinion:
– No way of reversing the replication after doing a fail over.
– No automatic removal of source VMs and claiming them on the DR site, which is very useful if you run a single vCenter.
– Claim only works for a single VM at a time.
– Multiple selection requires you to check each VM, one by one.

I talked with support and they registered IDEA-426 for the three last ones.
A feature for reversing the replication would be very nice, but face it. Number one priority will be to fail over to the DR site as quickly as possible. The source array might not even be there anymore… you didn’t just do the fail over for fun did you?
To check all the boxes for an Enterprise audit I do see the need to reverse the replication easily.

The end goal here would of course be VVol with synchronous replication and Peer Persistence for 3PAR/Nimble. At least for sites with relatively short distance between them. Sadly I don’t think I will see that in 2019, and probably not in 2020 either. 2021 maybe? I will continue to request it every time I talk with VMware and HPE about this 🙂

I might publish the SPBM DR scripts at a later stage, but I’m not done tweaking them, and the guys at HPE said they will probably release an updated version in the future.

NimOS was 5.0.6 when I wrote this.

PS; Cody Hosterman at Pure Storage has a great article covering VVols and replication


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s