Uploaded image for project: 'Product Roadmaps'
  1. Product Roadmaps
  2. MMF-1766

Detailed feedback for failed analyses


    • Type: MMF
    • Status: Closed
    • Priority: Major
    • Resolution: Fixed
    • Fix Version/s: None
    • Labels:



      With MMF-1762 SonarCloud shows visual feedback for ongoing analyses. But when analysis failed, users only get a generic message with no explanations about what happened and how to resolve the issue:

      The last analysis has failed. If it keeps on failing after several attempts, please contact us on Community forum and provide the following failing analysis ID: “xxxxxxxxxxxxxxx”.

      We see more and more of those messages on Community, and it is not sustainable to investigate all of them. We want users to know what is going on in case of failure, and provide as much as possible an actionnable message to help them fix the issue on their own.


      As soon as an AutoScan analysis is failing, I expect SonarCloud to show me relevant feedback. Error messages will explain briefly what is the source of the problem, and give relevant advice for troubleshooting. When there is a need for more explanations or many troubleshooting solutions possible, we will link to a relevant page of documentation where the user will find more details about the problem and how to solve it.

      Documentation Page

      A new page named Troubleshooting AutoScan will be created. Each possible type of errors will be documented in a section. Anchor links to these sections can then be linked from the UI error messages. The different sections will be:

      • Functional errors:
        • Configuration error in your .sonarcloud.properties file
        • Maximum number of Lines of Code reached (for private projects)
          • Link to the billing documentation
          • The doc explains I have to make a new push on my GitHub repository after I edited my billing plan to trigger a new AutoScan analysis
      • Technical errors
        • OOM
        • Timeout
          • This page will explain that Timeout happens after 30mn. We will also explain here that sometimes GitHub repositories are too large to be cloned under the 30mn time limit.
        • Compute Engine
          • This page will explain what the CE is and why sometimes it cannot receive reports correctly.

      UI error messages

      Error messages will appear using the banner component that we already use for analysis in progress status.

      • Functional error. I expect the UI to give me a way to get more details about why it failed and how can I fix the problem
        • Configuration error (Ex: sonar.sources and sonar.tests overlap / exclusions for copy-pasted detection / non-UTF-8 encoding detected in .sonarcloud.properties file)
          • The UI tells me something is wrong in my configuration of sonarcloud.properties file
          • I see a link "More info and troubleshooting advice" that goes to the Troubleshooting AutoScan documentation, with a direct anchor to the .sonarcloud.properties section
        • Max LoC reached (for organizations under private plan)
          • The UI tells me I reached max LoC in my plan
          • If I have sufficient user permissions, I'm asked to edit my plan in Billing page with a link
          • If I don't have sufficient user permissions, I'm asked to contact org administrator


      • Technical error. I expect the UI to tell me that something wrong happened, that the analysis could not complete and that I should try again (and at some point tell the user to go ask on Community / Contact form in case the problem still happens). In that case, we expect to have on our side the monitoring of technical errors to be aware of such situations and have actions to fix them.
        • OOM, Timeout (Analysis couldn't complete after 30mn), Can't send report to Compute Engine
          • The UI tells me the analysis failed because of an OOM, Timeout or Compute Engine problem
          • I'm redirected to the Troubleshooting documentation page, with a direct anchor to the related technical error. The link should say "More info and troubleshooting advice"
        • Can't clone GitHub repository because it is too large and the cloning exceeds our 30mn timeout
          • The UI tells me the GH repository is too large to be cloned by AutoScan in a reasonable timeframe.
          • I'm suggested to try another analysis method instead: with a CI tool or manually
          • A link should take me to the first page of the tutorial, where I can select another analysis method.
        • Can't clone GitHub repository because of issues on GitHub side
          • The UI tells me something is wrong on GitHub side and I should retry analysis later by making a new push on my GitHub repository. "Something went wrong on GitHub side. Please retry analysis later by making a new push on your GitHub repository"
        • Background task failed
          • The UI tells me analysis failed because of "name of the background task"
          • If I have sufficient rights, I will see a link to the Background task page
          • If I don't have rights to see the Background Tasks page, the UI tells me to contact my org Administrator


      To be continued




            • Assignee:
              fabrice.bellingard Fabrice Bellingard
              laura.wacrenier Laura Wacrenier
            • Votes:
              1 Vote for this issue
              3 Start watching this issue


              • Created: