Qlik Replicate Disaster Recovery

[lguevara]

I have an escenario when I have a linux server with Qlik Replicate.

But in an case with disaster recovery I want replicate the server to another secondary server with all settings same (Full server replication).

Then,

Are there any limitations or constraints to this approach?

In this scenario, when failover occurs from the primary to the secondary server, does the replication process initiate automatically or is a manual script required to trigger it?

Thanks

[Dana_Baldwin · Accepted solution]

Hi @lguevara 

Please refer to this section of the User Guide and let us know if you have any follow up questions. The Replicate server process should start and all tasks that were running resume upon fail over:

Setting up Qlik Replicate in a cluster environment #Setting up Qlik Replicate in a cluster environme...

Thanks,

Dana

[lguevara]

Hi @Dana_Baldwin ,

Yes I review about HA.

I'm about to perform a full replication of a Qlik Replicate server to a standby server. Once a failover occurs, can I expect the standby server to automatically take over and function without the need for manual reconfiguration of licenses and endpoint passwords?

Thanks

[Dana_Baldwin · Accepted solution]

@lguevara 

Yes if you follow the installation steps in the user guide (you have to install Replicate on the second server rather than clone it - the data directory has to be only used by one node at a time) then on fail over the tasks that were running will resume unless an unexpected error occurs.

Hope this helps clarify.

Thanks,

Dana

[lguevara]

In a disaster recovery scenario where a Linux server is clone completely exactly to a secondary server using VMware SRM, and a failover occurs, will the cloned server function correctly upon startup, or will licensing, master key, other issues be encountered?

[Dana_Baldwin]

@lguevara I don't know - I suggest you work with our Professional Services team to design a solution that will meet your needs.

If the clone happens with any amount of time before the fail over, the data directory will be in a different state. You also will probably run into master key issues if you have more than one data directory - this has to be a shared resource between the two servers in a fail over cluster. If it's set up the way our documentation requires, you should not run into a master key issue as you will have already defined it to be usable between the two servers. If you have two data directories, one on each server, I don't see how that will work well - but this needs to be looked at with more detail at a project level to design the best solution. Please reach out to your Account Manager to get help from Professional Services on this.

Hope this helps,

Dana