Recently we have been facing delay issues in Sonar analysis of Community edition while scanning. Specifically whenever a large project is executed typically which would ideally take 5 minutes to get analyzed, we observe the delay in fetching the Quality gates. We are using SonarQube LTA version 9.9 to mitigate this issue we tried to
increase the resources(CPU and memory limit) for SonarQube
keep the optimum DB size
If the cause of the delay is due to single worker with a mono thread in the Community Edition because of which runs scans in a sequential manner. As a result, upcoming jobs are put on hold until the large project scan is completed.
Most of the time-taking jobs are typically large projects. Is there a way to resolve this issue for community edition. Does moving to latest LTA would be any help here?
In our case, we only update our productive instance to current LTA versions. Does moving latest LTA would help in such resolving delay issues OR should we specifically update to Community Build (v25.2)
We have upgraded to Sonar Community Build 25.4
Even after the upgrade, the delay in analysis still happens. For example, the problem occurs when a particular project runs its scan. Ideally the project takes below 10 minutes to complete its scan. Basically, it is .NET codebase. they are scanning only the code which is relevant, they have configured exclusions in their configuration file.
When we checked the logs, we see as below.
INFO ce[o.s.c.t.s.ComputationStepExecutor] Software composition analysis unavailable | status=SUCCESS | time=0ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Checks executed before computation of measures | status=SUCCESS | time=0ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Generate SQ Upgrade analysis events | status=SUCCESS | time=2ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Load file hashes and statuses | status=SUCCESS | time=596ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Load Quality gate | status=SUCCESS | time=4ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Load new code period | status=SUCCESS | time=2ms
2025.06.19 03:19:44 INFO ce[o.s.c.t.s.ComputationStepExecutor] Load prioritized rules | status=SUCCESS | time=5ms
followerd by series of Warnings
/api/user_groups/search Web service is deprecated since 10.4 and will be removed in a future version.
and some more warnings on
2025.06.19 04:02:46 WARN es[o.e.d.x.ParseField] data_stream.dataset=“deprecation.elasticsearch” data_stream.namespace=“default” data_stream.type=“logs” elasticsearch.event.category=“api” event.code=“deprecated_field_from” message=“Deprecated field [from] used, this field is unused and will be removed entirely”
With all these warnings and scan which ideally finishes before 10 minutes, runs for 3-4 hours. Due to which other jobs are also getting affected and they are waiting in queue in ‘PENDING’ state.
Can you please suggest how can we reduce such issues.
There are probably a few red herrings there (I know that Docker likes to put all the logs in one place).
What Database do you have backing up your SonarQube server? Oracle? Postgres? MSSQL?
We document some good practices for keeping your database performant (especially after upgrades). The Compute Engine (what is processing SonarQube analyses) is heavily reliant on DB performance.
Only the latest version of SonarQube Community Build is considered active, so you’ll need to update and see if the situation is still replicable before we can help you.
Only the latest version of SonarQube Community Build is considered active, so you’ll need to update and see if the situation is still replicable before we can help you.