Yellow Bricks

by Duncan Epping

Demo time – vCloud Director 5.1 disaster recovery demo

· ·

When I was playing with the new vCloud Director 5.1 and Site Recovery Manager 5.1 I figured I would record a demo of the DR solution that Chris Colotti and I developed. The demo is fairly straight forward and hopefully helps you in the process of building a resilient cloud infrastructure. In this demo I have included:

  • vSphere 5.1
    • vSphere Replication
  • vCloud Director 5.1
  • Site Recovery Manager 5.1

Reader Interactions

Comments

  1. Hi,

    Thank you for preparing this demo video. It nicely illustrates the approach described in the white paper you co-authored with Chris Colotti.

    Have the new version of the various products involved (SRM, vCD) helped simplify the configuration at all? Or is vCD resiliency essentially the same as it was with vCD 1.5 and SRM 5.0?

  2. Hi Duncan,

    Thank you for this video.
    Can you post your command script used in step 10?

    I have to set up such a vCloud Director DR with SRM, both 5.1
    It’s really hard to get Informations about this from VMware, like best practises etc.

    Greets

    Oliver

    • Sorry, I can’t share this script. But I know VMware PSO has been involved with various engagements and created custom scripts for the solutions you mention. I suggest contacting local PSO resources.

  4. Hi Duncan.

    I have a doubt.

    I have a Site A in my building office with 3 servers hosting 38 VM in vSphere 4.1 and vCenter Server, then like to replicate to a Site B taht will be hosted in my country Datacenter.

    I like to get a solution Recovery and Failover, i know about SRM and know that need a licence, so there is a Datacenter solution for SRM, provider talk about a licence in both Sites, because there is not a SRM for here full Datacenet VMWare Environment.

    That is true ?

    Thanks

  5. Thanks for posting this. It helps in some cases. For myself I need to be able to provide a more geographically dispersed DR site, so stretching a cluster with hosts in maint mode would not be option I would think. Did you intend this to be a local/metro range solution? WAN Opt (Riverbed) may help stretch this a little further too.

    • The fact that the “cluster” is stretched is just to make recovery easier, not because there is a need for a true stretched cluster. You could also have the hosts outside of the cluster and move them in when new resources are required.