cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Looking to upgrade from Dynatrace Managed to SaaS? See how

How to setup disaster recovery site for Dynatrace Managed

ram16
Helper

Hi All,

Currently we have Dynatrace managed with 3 cluster node. Now, we need to setup disaster recovery site with one cluster node or more nodes. How do we set it up? Dynatrace documentation is little confusing in this topic.

Ramanan Raghunathan
3 REPLIES 3

Radoslaw_Szulgo
Inactive

Hi @ram16 , 

With 3 nodes, you're already highly available - meaning you can survive the loss of one node. If you add two additional nodes having 5 in total - you can afford to lose up to 2 nodes in the cluster.

You can also achieve additional fault tolerance by redundancy, splitting the cluster into:

- separate racks - https://www.dynatrace.com/support/help/setup-and-configuration/dynatrace-managed/fault-domain-awaren...

- and/or data centers - https://www.dynatrace.com/support/help/setup-and-configuration/dynatrace-managed/fault-domain-awaren...

In case you want to set up a disaster recovery procedure for the situation of losing more than that, you'd need to look at the backup/restore procedure: https://www.dynatrace.com/support/help/setup-and-configuration/dynatrace-managed/operation/back-up-a...

 

Hope this helps to start!

Senior Product Manager,
Dynatrace Managed expert

Hi Radoslaw,

Thanks for the details. As per the local government rule we need to shutdown the servers from Head office and run the operation in Disaster recovery site for 1 month. As per the documentation 1 node can go down with 3 cluster setup. 2 nodes can go down in 5 cluster setup. If this is the case, then i need to setup 6 cluster node setup.


So that i can keep 3-3 in each location and bring down 3 nodes during our DR activity. But i am thinking why should we add 3 more nodes to cluster in the DR site which will run only for 1 month. Because our environment is very small. Just thinking how best I can do to maintain the High availability of the cluster as well as to reduce the hardware cost.

Ramanan Raghunathan

Commenting for continuous discussion

Ramanan Raghunathan

Featured Posts