Uploaded image for project: 'Bigtop'
  1. Bigtop
  2. BIGTOP-961

hcatalog should be re-using hive user when running

    Details

    • Type: Improvement
    • Status: Closed
    • Priority: Blocker
    • Resolution: Fixed
    • Affects Version/s: 0.6.0
    • Fix Version/s: 0.6.0
    • Component/s: deployment, general
    • Labels:
      None

      Description

      Given that hcatalog and hive actually share tons of common metadata it makes sense to run both under the same uder id (partially to solve legacy issues with existing hive deployments)

        Activity

        Hide
        mgrover Mark Grover added a comment -

        Roman, would it make sense to use a separate hcatalog user that's a part of the hive group instead?

        Show
        mgrover Mark Grover added a comment - Roman, would it make sense to use a separate hcatalog user that's a part of the hive group instead?
        Hide
        anatoli.fomenko Anatoli Fomenko added a comment -

        +1

        Show
        anatoli.fomenko Anatoli Fomenko added a comment - +1
        Hide
        rvs Roman Shaposhnik added a comment -

        Mark Grover that doesn't seem to work as far as accessing common metadata is concerned. Also, on a more philosophical note, I guess I'd say that now that HCatalog is part of Hive we should probably treat it as such, would you agree?

        Show
        rvs Roman Shaposhnik added a comment - Mark Grover that doesn't seem to work as far as accessing common metadata is concerned. Also, on a more philosophical note, I guess I'd say that now that HCatalog is part of Hive we should probably treat it as such, would you agree?
        Hide
        mgrover Mark Grover added a comment -

        Hi Roman, good point. I think for now it's ok for us to change the user to hive user. Having 2 separate users would have puts some restrictions on the permissions of hive warehouse directory on HDFS and the embedded derby file based storage on local filesystem at the very least. Changing to hive user should fix those.

        I have been thinking about this a little more and I think moving forward, it would be really nice to have a standalone RDBMS in Bigtop which we can use to house all metastores (hive, oozie, hue, etc.) in Bigtop. I will create a separate JIRA (for Bigtop 0.7) for that but when that happens, perhaps it would make sense to revert back to HCat user since we would no longer be using embedded derby and can easily manage the permissions of hive warehouse directory on HDFS.

        You have my +1 (non-committer).

        Show
        mgrover Mark Grover added a comment - Hi Roman, good point. I think for now it's ok for us to change the user to hive user. Having 2 separate users would have puts some restrictions on the permissions of hive warehouse directory on HDFS and the embedded derby file based storage on local filesystem at the very least. Changing to hive user should fix those. I have been thinking about this a little more and I think moving forward, it would be really nice to have a standalone RDBMS in Bigtop which we can use to house all metastores (hive, oozie, hue, etc.) in Bigtop. I will create a separate JIRA (for Bigtop 0.7) for that but when that happens, perhaps it would make sense to revert back to HCat user since we would no longer be using embedded derby and can easily manage the permissions of hive warehouse directory on HDFS. You have my +1 (non-committer).

          People

          • Assignee:
            rvs Roman Shaposhnik
            Reporter:
            rvs Roman Shaposhnik
          • Votes:
            0 Vote for this issue
            Watchers:
            4 Start watching this issue

            Dates

            • Created:
              Updated:
              Resolved:

              Development