Zerto for Disaster Recovery

Zerto for Disaster Recovery

December 2, 2013 2 By Eric Shanks

ZertoVMworld

 

As many of you know I’ve been a fan of the VMware Product called Site Recovery Manager (SRM) for a disaster recovery solution, I even wrote an SRM 5.5 Guide for using it.

But many people I talked to told me to check out Zerto as a DR solution because of how simple it was to use and setup.  I figured that I owed it to them to at least try them out, and so they are now in my lab.

Disclaimer- As of the time of this writing Zerto is a sponsor of theITHollow.com but did not pay for any reviews or posts based on their software.

 

My first thoughts about the Zerto solution really was how easy it was to get up and running.  It required the Zerto Virtual Manager to be installed on a windows instance for each of my sites.  The managers can then deploy a Zerto Replication Appliance (Z-VRA) for each ESXi host in my site.  The Z-VRA does the all of the heavy replication workload so it’s a good way to scale out your DR solution if the protection site needs to grow.  Needing to add more hosts and more VMs, just add more Z-VRAs.

Deploying the Z-VRAs is a matter of clicking Install New VRA and giving them an IP Address.

ZVRA

 

Below is a graphic that explains the traditional Site to Site DR scenario.  You can see a ZVM and the Z-VRAs and how a single ZVM manages multiple replication appliances.  The important thing to notice is that the type of storage does not matter.  In my lab I’ve used a Synology NAS on the production side and local storage on the DR Side.

zertoarchitecture

 

Once your Servers are up and running, the next step is to pair the sites.  This involves adding the name or IP address of the opposite site, and when done the pairing is completed.  No need to redo this operation on the adjacent side.

From here, you need to create a VPG.  This Protection Group maps the Replicated VMs to the appropriate datastores and networks on the replication site, as well as the targeted Recovery Point Objective (RPO)  Also, it’s important to notice that both a Failover network and a “Test” failover network can be specified so that test can be done without affecting production.  Lastly, configuring the Virtual Machines that should be replicated, is a matter of adding the VM to a VPG.

VPG

 

ZertoVMReplication

The next part is difficult…Waiting.  Now the VM will start the process of replicating their data to the secondary site.  Let this process finish and the Zerto Manager will give a really simple to view graphical representation of your DR Solution.  I really love the simple interface.  Just from looking at the GUI it’s very simple to see the direction of the replication as well as the performance of the replication.  And a great big button ready for failovers.

ZertoSites

Zerto allows you to test the failovers, run real failovers and create offsite clones which is a nice feature.  Also during failover operations multiple point in time selections can be selected so the “Last Replication” doesn’t have to be used if not wanted.

When running a “Read” failover, click the giant failover button and follow a couple of quick prompts.  First, select the VPG you are looking to failover.

ZertoFailover1

Next, select the checkpoint (multiple point in time replication) and what to do with the Production VM, that is if it still exists after the disaster.  Also, notice there is a commit policy which is a neat addition.  Here I’ve set it to automatically commit the failover after 10 minutes.

ZertoFailover2

Lastly, a big red button warning will show up to make sure you REALLY REALLY REALLY want to do this.  “Seriously, you want to declare a disaster?  Really????

ZertoFailover3

Notice that once the failover is done, a countdown timer showed up for me counting down from 10 minutes.  This is from the commit option I selected earlier.  Here I just pressed the “commit” button to go ahead and commit my changes.

ZertoFailover4

A cool thing happened as soon as I committed my failover.  An option to reprotect the VM was prompted.  How cool is that?  If the original Production site is available the Replication will reverse directions to get you ready to fail back.

ZertoFailover5

Zerto has a variety of other features that I may cover in another post, such as the ability to protect a site with the peer site being a public cloud network.  http://www.zerto.com/bcdr-for-cloud-providers/dr-to-the-cloud/

If you’d like to know more about this solution, please check them out for yourself and grab a free trial while you’re at it.