Qlik Community

Qlik Sense Deployment & Management

Discussion board where members learn more about Qlik Sense Installation, Deployment and Management.

Announcements
See why Qlik is recognized as a Leader for the 10th year in a row – and discover how Qlik can help put your business in the lead. Get Report
Partner
Partner

Repository DB setup for Failover candidate Qliksense

Dear All,

I am confused about how to set up a repository database for Qliksense cluster deployment with failover candidate.

1.how we can set up the repository DB? 

2. do we need a separate Postgres SQL DB server in order to implement the automatic failover?

3. do we need SAN/NAS storage for share folder?

kindly help me understand this.

thanks,

Bhasker

 

 

1 Solution

Accepted Solutions
Highlighted
Employee
Employee

Re: Repository DB setup for Failover candidate Qliksense

> 1.how we can set up the repository DB? 

https://help.qlik.com/en-US/sense-admin/September2019/Subsystems/DeployAdministerQSE/Content/Sense_D... is one approach.

> 2. do we need a separate Postgres SQL DB server in order to implement the automatic failover?

No, but failover when the PostgreSQL database is hosted on one of the central nodes doesn't get you much.

3. do we need SAN/NAS storage for share folder?

No, but hosting the share folder on one of the central nodes doesn't get you much.

Reference video: https://www.youtube.com/watch?v=Y5IblMGwrLU

 

Let's use this as a reference point:

failover.png

With this architecture, it is entirely possible to host the Repository database and/or file system on the Active Central. But this undermines the purpose of failover. The purpose of failover is that if the active node goes down then the passive will take over responsibilities. With the Repository DB and file system on one of the servers, if the server goes down then the entire site is down.

View solution in original post

1 Reply
Highlighted
Employee
Employee

Re: Repository DB setup for Failover candidate Qliksense

> 1.how we can set up the repository DB? 

https://help.qlik.com/en-US/sense-admin/September2019/Subsystems/DeployAdministerQSE/Content/Sense_D... is one approach.

> 2. do we need a separate Postgres SQL DB server in order to implement the automatic failover?

No, but failover when the PostgreSQL database is hosted on one of the central nodes doesn't get you much.

3. do we need SAN/NAS storage for share folder?

No, but hosting the share folder on one of the central nodes doesn't get you much.

Reference video: https://www.youtube.com/watch?v=Y5IblMGwrLU

 

Let's use this as a reference point:

failover.png

With this architecture, it is entirely possible to host the Repository database and/or file system on the Active Central. But this undermines the purpose of failover. The purpose of failover is that if the active node goes down then the passive will take over responsibilities. With the Repository DB and file system on one of the servers, if the server goes down then the entire site is down.

View solution in original post