I've setup a simple chkdsk procedure which outputs the results to a txt file, i then have a testFile step to look for errors and write the error in the procedure log. I found that the results on many agents are inconsistent with the results i'd get when doing a manual chkdsk scans. I've tested a few agents, my manual scan consistently finds no errors compared to the output results product by the procedure which consistently shows errors.
Is the procedure running the shellcommands different than when running it manually within the OS? Running the shellcommand as User doesn't change anything.