Uploaded image for project: 'FishEye'
  1. FishEye
  2. FE-1988

Allow a scan agent to complete the index of a repository on another server and copy over the cache once it is done to reduce the load on the main server.



    • Our product teams collect and evaluate feedback from a number of different sources. To learn more about how we use customer feedback in the planning process, check out our new feature policy.


      Pros for having a "Scanning" Agent approach:

      • Scaling Fisheye becomes a matter of adding an agent to scan Source Code on new hardware
      • Clients who have very large code bases can provide their own hardware to help off-set the cost of the Fisheye plant
      • Adding a new Fisheye repository will not cause all other users to suffer slow downs due to the CPU usage being used up by the scanning

      From https://support.atlassian.com/browse/FSH-2026

      Current workaround is (which has to be done manually:

      Set up a test server and add the repository there, and wait for it to finish the initial scan there. Once this is finished,

      1) Shutdown test server
      2) Add the repository to your production server via Administration > View Repository List > Add repo but set "Enable Immediately" to "No".
      3) Copy the FISHEYE_INST/var/cache/ADDEDREPONAME directory, FISHEYE_INST/var/data/DiffTextCache/ADDEDREPONAME directory to the production server's FISHEYE_INST/var/cache and {{FISHEYE_INST/var/data/DiffTextCache directory respectively.
      4) Go to Administration > View Repository List > Enable, then Start next to the repository.

      This should save your production server the hassle of having to do the initial scan for the repository and the entire load is taken by your test server.




            Unassigned Unassigned
            pkamal Partha
            2 Vote for this issue
            5 Start watching this issue