Uploaded image for project: 'Hadoop YARN'
  1. Hadoop YARN
  2. YARN-4904 YARN Log tooling enhancements
  3. YARN-5312

Parameter 'size' in the webservices "/containerlogs/$containerid/$filename" and in AHSWebServices is semantically confusing

    XMLWordPrintableJSON

Details

    • Sub-task
    • Status: Open
    • Major
    • Resolution: Unresolved
    • None
    • None
    • None
    • None

    Description

      This got added in YARN-5088 and I found this while reviewing YARN-5224.

      Also, the parameter 'size' in the API "/containerlogs/$containerid/$filename" and similarly in AHSWebServices is confusing with semantics. I think we are better off with an offset and size.

      An offset (in bytes, +ve to indicate from the start and -ve to indicate from the end) together with a size (in bytes) indicating how much to read from the offset are a better combination - this is how most file-system APIs look like, for comparison.

      I can also imagine number of lines as a better unit than bytes for offset and size - perhaps yet another ticket.

      /cc vvasudev.

      Attachments

        Issue Links

          Activity

            People

              vinodkv Vinod Kumar Vavilapalli
              vinodkv Vinod Kumar Vavilapalli
              Votes:
              0 Vote for this issue
              Watchers:
              4 Start watching this issue

              Dates

                Created:
                Updated: