Is the central node in multi node deployment a single point of failure?

[kausar007]

Hi Guys,

I am planning a deployment of highly available Qlik Sense server. Reading through documentation I found out that central node is required for all kind of deployments and also "Rim nodes synchronize data through the central node.". How is not the central node then single point of failure? What happens if the central node goes down?

Kind Regards,

Kausar

[jaisoni_trp]

You can make another node as fail over candidate and swap Central node in Share Persistence. I would suggest to move to Share architecture instead of Synchronized.

[kausar007]

Hi Jai,

Thanks for the suggestions. Do you have any tutorial/guide links for how to install Qlik sense components as shared architecture with high availability?

Regards,

Kausar

[jaisoni_trp]

Qlik is doing a good job documenting everything on help site. I would start with that.

http://help.qlik.com/en-US/sense/September2017/Subsystems/PlanningQlikSenseDeployments/Content/Deplo...

[kausar007]

Hi Jai,

Yes I was reading through this documentation that means I am on the right track and I will be using the shared persistence option and not the synchronized. As an example let's say I am creating something like this:

I have a central node and two proxy/engine nodes But still my question is what happens if the central node goes down?

Thanks,

Kausar

[jaisoni_trp · Accepted solution]

Here you go

Failover ‒ Qlik Sense

[kausar007]

Hi Jai,

This is great. Thank you for the link. I was looking at the version 3.2 documentation as I had 3.2 installed. Looks like failover was introduced later. I will be looking to upgrade now.

Regards,

Kausar