When using the clang tool for parsing warnings in the console of a parallel pipeline (10 build stages which all generate warnings) the number of warnings are always summed up. It seems that the stage scans always the full console at the point of execution.

Note: I am using a scripted pipeline.

Is this a known issue or am I doing something wrong?

Example code (project_x / project_x_name is different for every parallel stage):

def clangIssues= scanForIssues id: "${project_x}", name: "${project_x_name}",  tool: clang(id: "project_x", name: "project_x_name"), aggregateResults: false
publishIssues: issues: [clangIssues], id: "${project_x}", name: "${project_x_name}"