Solved

Does AHV support Automatic Site Failover or supports Metro Witness?

  • 2 April 2020
  • 3 replies
  • 1802 views

I have read that it is only supported using ESXI. If so, what are the other ways to automate the failover.

icon

Best answer by JeremyJ 2 May 2020, 01:52

View original

This topic has been closed for comments

3 replies

Userlevel 4
Badge +5

Hello @cviernes-59806  

Based on the Metro Availability Guide
https://portal.nutanix.com/page/documents/solutions/details/?targetId=BP-2009_Metro_Availability:BP-2009_Metro_Availability

“Metro Availability does not currently support AHV environments.”
But I do remember it being in the pipeline and soon will be supported in the future release of AOS and AHV. 

You can refer to the following guide to configure Async DR in AHV environment  
https://portal.nutanix.com/page/documents/solutions/details/?targetId=BP-2005_Data_Protection:BP-2005_Data_Protection
 

Hello @cviernes-59806  

Based on the Metro Availability Guide
https://portal.nutanix.com/page/documents/solutions/details/?targetId=BP-2009_Metro_Availability:BP-2009_Metro_Availability

“Metro Availability does not currently support AHV environments.”
But I do remember it being in the pipeline and soon will be supported in the future release of AOS and AHV. 

You can refer to the following guide to configure Async DR in AHV environment  
https://portal.nutanix.com/page/documents/solutions/details/?targetId=BP-2005_Data_Protection:BP-2005_Data_Protection
 

Can I still do automatic failover using Async DR?

Userlevel 3
Badge +4

This isn’t currently an available feature. If and when it becomes available I would expect to see it covered in the Data Protection and Disaster Recovery Tech Note or Best Practices Guide.

With synchronous replication it makes a lot more sense to have an automated fail-over.

With async-DR there is necessarily a decision point for most environments between restoring to your last snapshot and giving up any newer data, or working to restore function at the primary site, possibly at the cost of further downtime.

If you have a business case for witness-driven Async-DR failover which would not be adequately met by metro-availability on AHV this would be worth logging a case to get the request for enhancement process started. If synchronous replication (metro-availability) on AHV with witness-automated failover would meet the need, I’m fairly certain that will be available sooner than the completion timeframe of a new RFE, but again if you have a use-case for it the RFE could be worth filing.