Failover Testing with One-to-Many Replication

Hello,

Bar from Zerto here.

You can find information relevant to your question in our documentation:
http://s3.amazonaws.com/zertodownload_docs/Latest/Zerto%20Virtual%20Manager%20vSphere%20Administration%20Guide.pdf

Page 357 at the bottom provides information about “Reverse Protection with One-to-Many”.

Hello Bar,

Thanks for the information. I am confused by this specific point in the documentation you sent me regarding resuming VPG replication after reverse replication has been initiated from a recovery site to a protected site:

To resume replication do the following:
1. Open the Edit VPG wizard.
2. Select the virtual machine that was previously deleted from the protected site.
3. Remove the virtual machine from the VPG.

Why would the VPG need to be removed? Wouldn’t the VPG pick up the new VM that was just reverse replicated to the protected site as the VM that should still be replicated in one-to-many VPGs?

I am looking at this section for reference:

Reverse Protection with One-to-Many
Zerto enables you to protect virtual machines in a maximum of three VPGs. These VPGs cannot be recovered to the same site.
If Reverse Protection is selected, and the virtual machines are already protected in other VPGs, the following occurs:
■ The virtual machines are deleted from the protected site.
■ The replication of other VPGs containing these machines is paused.
■ The recovery of all the virtual machines, including the virtual machine removed from the protected site, is possible.
To resume replication do the following:
1. Open the Edit VPG wizard.
2. Select the virtual machine that was previously deleted from the protected site.
3. Remove the virtual machine from the VPG

Thanks for the information.

Hello,
Bar from Zerto here.

Taking an example of 1 Prod (A) site replicating to 2 DR (B + C) sites,
VPG 001 Prod A > DR B
VPG 002 Prod A > DR C

Once an administrator chooses to failover live >commit>reverse protection VPG 001, VPG 002 will enter pause state.

VPG001 Reverse protection deletes the VM on site Prod A as uses the VM disks as preseed.

Hence VPG001 will enter delta sync to Reverse protect as configured, yet VPG002 is configured to protect a VM that resides on Prod A.
That’s why VPG002 will enter pause state as it would not “pick up” the new VM just created on DR B.