Background task time different with same project but different instances

SonarQube Server / Community Build
1
  • Sonarqube 10.2.1.78527*
  • Deployed into OpenShift cluster with helm chart 10.2.1*
  • Goal - decrease background task execution time*
  • what have you tried so far to achieve this - update sonarqube, add more resources to the instance*

Hello. I have two equal instances of sonarqube deployed into 2 same cluster. I’ve imported one project into both instances, and integrated scan task into cicd pipeline. One test branch has approx 1 commit per day and during scan background task take approx 1-2 minute it scanned by instance A, another branch has 5-10 commits per day but during scan background task takes more than 10 minutes that leads to the error “Quality Gate status timeout”. Of course I can fix it with changing timeout but this will extend pipeline execution time
Plugin that used sonar-maven-plugin:3.11.0.3922
What may affect background task execution time and how can I reduce it? Project has about 400k lines of code

2

Hey there.

The most likely culprits for compute engine performance are latency between the database and SonarQube, database performance (CPU / Memory / Disk), and regular database maintenance.

What database are you using (Postgres, SQL Server, Oracle) and where is it located relative to your SonarQube instance?

3

It’s postgres, deployed in the same cluster and same namespace, I used helm chart for the deployment so it manages both instances, also as I see that cpu and memory consumption is minimal, like 200milicores vs 1G and requests are 1CPU vs 2Gb. About database maintenance, did I miss something, I thought it provided by ttl or something like this, and during updates it’s reindexing, maybe I need some extra things to do?

4

I’m not really sure what you’re referring to by ttl.

The maintenance I’m referring to is essentially what’s outlined in this post:

5

Thank you very much for your answer. Actually I didn’t try this before. Under TTL I meant only housekeeping settings in administration panel. But today I’ve made something similar according this https://docs.sonarsource.com/sonarqube/latest/setup-and-upgrade/upgrade-the-server/post-upgrade-steps/#clean-up-postgresql
Hope it will help. I thought such tasks executed automatically during updates, but as I understood not

1 Like
6

Finally my issue has been resolved with es reindexing. I’ve deleted /data/es8 folder and restarted contained. Background task time become normal. Hope won’t need to do it often :slight_smile: