Must-share information (formatted with Markdown):
- which versions are you using (SonarQube Server / Community Build, Scanner, Plugin, and any relevant extension): SonarQube 9.9.0.65466,
- how is SonarQube deployed: zip
Hey, a long time ago I set up our C++ scanning to work off an analysis cache not long after that feature was added. To achieve this, I had to:
- In our Azure Devops CI pipeline
- copy the source directory from its default checkout location (C:\agent_work?\s\ where ? would be a random number depending on the build machine) to a static location (C:\static-agent)
- build
- Cache the output to our pipeline
- In our Azure Devops PR pipeline
- Grab the cache from the CI pipeline, using TargetBranch property to find the right one
- copy the source directory from its default checkout location to the static directory, to ensure it matches
Im doing some updates at the minute and wanted to revisit this. I see there is a server cache option now also, and have a couple questions:
- The docs say to use filesystem (as we are today) if
The server cache size becomes a concern.
Is this a storage question for us, or is there some recommendation that if the caches are over X size, use filesystem?
- Is moving the files to a static location still a requirement for filesystem implementation, so the filepaths always match across a build agent pool? Is this also a requirement if using server cache?
thanks for any advice