Step Away From Array – A Disaster Recovery Story
Last week, we sent out an email entitled, “Step Away from Array” detailing the advantages of Zerto Virtual Replication over Array-based replication solutions. One response, from @roidude on twitter, touched off my idea for today’s blog post:
(My apologies to Dr. Seuss…)
So we sat in the house with our old, wet Array.
Array said, “Look at me now! Look at me go!”
“I replicate with your storage. I have short RTOs!”
What about Storage sprawl?
“Won’t do any harm.”
“Just build up an entire duplicate server farm!”
Please go away, we started to beg,
He said, “But I already cost you an arm and a leg.”
Can you recover just one specific VM?
“No. Ask VMWare for that, we leave that to them.”
And then he pulled out a box, marked Thing Two and Thing One,
“SRM and vSphere Replication add so much fun!”
And that is when things started to look like a zoo:
We located VMs affecting our app,
Then documented all LUNs, not forgetting to map.
Next moved all other app VMs to a separate LUN,
Now test and check — yes, this is how it’s done.
Configure replication pairs, allocate more LUNs, then verify install.
Make sure replica has the same properties, IPs and all.
vMotion’s a problem, vSphere versions are, too.
They said, “Replication’s not supposed to be easy for you!”
And I said, “I do not like the way they play!”
If Mother could see this, Oh, what would she say!
As fast as you can, think of something to do!
Things are getting more complicated with Thing One and Thing Two!”
So Sally and I, we decided that day.
We finally said, “Step away from Array!”