Why pod sonarqube-postgresql-0 is not programmed in labelled node and which has taint and toleration

SonarQube
1

I am writing to ask for your support, I made a SonarQube deployment, I labeled the node, configured taint and toleration (as in the documentation), all good with the pod sonarqube-sonarqube-0, but with the pod sonarqube-postgresql-0 does not happen the same, it falls on any node in the cluster.
Could someone guide me in this issue?

The EKS node ip-10-70-51-142.ec2.internal is in NoSchedule mode.

k -n sonarqube get po -o wide
NAME                             READY   STATUS    RESTARTS   AGE   IP             NODE                           NOMINATED NODE   READINESS GATES
sonarqube-preprod-postgresql-0   1/1     Running   0          29m   10.70.52.31    ip-10-70-51-177.ec2.internal   <none>           <none>
sonarqube-preprod-sonarqube-0    1/1     Running   0          53m   10.70.63.225   ip-10-70-51-142.ec2.internal   <none>           <none>
2

Hi,

Sorry, but I’m not understanding the question.

What edition of SonarQube are you using?

Are you trying to run multiple, redundant nodes at the same time?

 
Ann

3

Thank you G Ann Campbell for replying, I am using Community 9.5.0 version.
According to the documentation: “We recommend binding SonarQube to a specific node and reserving this node for SonarQube”. I did all this and it works fine, but only for the sonarqube pod, not for the postgresql pod.

4

Hi,

That recommendation is about SonarQube. It’s not clear to me why you would apply the same to your Postrgres node. For Postgres, you should follow whatever the Postgres recommendations are.

 
Ann

5

OK, thank you for your reply

6

In production we move the database to an RDS or AuroraDB PostgreSQL instance, so this query no longer applies.
Thank you very much.