Differences on new code measures and overall code measures between analysis of the same code base


  • SonarQube server Enterprise edition
  • Build Wrapper for Windows from {SonarQube URL}/static/cpp/build-wrapper-win-x86.zip
  • SonarScanner

What are you trying to achieve:

I am trying to analyse part of big mono repository as SonarQube project. Part of the code base is huge enough to contain over 1 million of lines and mostly C/C++ is being used. After first successful analysis I have problem that every time I run analysis over same code base different amount of code smells and bugs are reported for both new code and overall code. New code setting for this particular project is:
Previous version: The New Code will be based on the analysis following the previous version.
There are some analysis warnings about file encoding and missing blame information. However, differences between analysis aren’t on the files with these problems.

Analysis warnings:

  • There are problems with file encoding in the source code. Please check the scanner logs for more details.
  • Missing blame information for 8 files. This may lead to some features not working correctly. Please check the analysis logs and refer to the documentation.

Tried so far:

  • SonarQube project recreation
  • making small change in code base between the builds to ensure new commit hash (like adding a single comment)
  • sonar.projectVersion property value’s incrementation between the builds
  • analysing different part of mono repository

Did anyone face something similar? Should new code’s options be somehow adjusted? Any help will be really appreciated.