- Access exclusive content
- Connect with peers
- Share your expertise
- Find support resources
Our community just got stronger! Explore new identity security resources, connect with experts, and discover more ways to learn and collaborate.
09-05-2018 02:31 AM
Can someone please share their experience of setting up a HA cluster with DR enabled?
If a client has two datacentres, do you install one Vault in each datacentre or do you put two Vaults in each data centre (2 clusters) with one cluster effectively being disaster recovery?
09-06-2018 06:57 AM
There are a number of factors but most times we look at the RTO/RPO the customer is expecting. Also we look at cost because in many cases, HA means additional hardware in the form of server + SAN. This causes cost to go up in a competitive situation. I’ll tend to go conservative and check if a prod vault + DR vault will serve the customer’s requirement before. If there are clear about it then i’ll not suggest HA altogether. If we throw in what goes up during an upgrade, the complexity maintaining it just increases.
That being said, Prod HA and DR HA does give the highest level of assurance for the customer and I have customers on that configuration without any problems and are pretty happy with it.
09-05-2018 02:59 AM
This depends on the duration of the DR.
i have seen implementation of HA cluster both for production and DR.
HA Production Vault Cluster
HA DR Vault Cluster
This was on the reason should there be an extended duration of a DR, the customer wanted to ensure there is no single point of failure since it has already failed over to the DR vault. Meaning if the ONLY active DR Vault fail then everything else will fail.
In short, you might want to fit the design based on the Recovery Time Objective expectation to design whether there is a need for a HA DR Vault Cluster.
Hope this help
09-05-2018 08:27 PM
@L a ,
Most important, if you choose for a single DR server that the hardware specification is the same as the Production nodes. And periodically check if the hardware is still sufficient. Companies do tend to grow.
And if you add hardware to the production nodes, don't forget the DR node(s)
09-06-2018 06:57 AM
There are a number of factors but most times we look at the RTO/RPO the customer is expecting. Also we look at cost because in many cases, HA means additional hardware in the form of server + SAN. This causes cost to go up in a competitive situation. I’ll tend to go conservative and check if a prod vault + DR vault will serve the customer’s requirement before. If there are clear about it then i’ll not suggest HA altogether. If we throw in what goes up during an upgrade, the complexity maintaining it just increases.
That being said, Prod HA and DR HA does give the highest level of assurance for the customer and I have customers on that configuration without any problems and are pretty happy with it.
09-07-2018 08:59 AM
Thanks for the responses guys.
09-10-2018 04:29 AM
Cost aside, I think a HA Prod and HA DR gives the highest level of redundancy and a very high RTO/RPO. However, split brain (Network issue at Prod Data Centre causes failover to HA DR, clients in part of the world talk to HA DR, and others to HA Prod) is still quite possible to occur. Your CA Admins have to be on the ball and understand the consequences of that type of scenario.
Click Accept as Solution to acknowledge that the answer to your question has been provided.
The button appears next to the replies on topics you’ve started. The member who gave the solution and all future visitors to this topic will appreciate it!
These simple actions take just seconds of your time, but go a long way in showing appreciation for community members and the LIVEcommunity as a whole!
The LIVEcommunity thanks you for your participation!

