Flume
  1. Flume
  2. FLUME-1653

Update Hadoop-23 profile to point to hadoop-2 alpha artifacts

    Details

    • Type: Bug Bug
    • Status: Resolved
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: v1.2.0
    • Fix Version/s: v1.3.0
    • Component/s: Build
    • Labels:
      None
    1. FLUME-1653.patch
      3 kB
      Hari Shreedharan

      Activity

      Hide
      Hari Shreedharan added a comment -

      Updating hadoop-23 profile to point to hadoop-2.0.0-alpha

      Show
      Hari Shreedharan added a comment - Updating hadoop-23 profile to point to hadoop-2.0.0-alpha
      Hide
      Brock Noland added a comment -

      Should we change it's name to hadoop-2? Also I think 2.0.2-alpha is out: http://www.apache.org/dist/hadoop/common/

      Show
      Brock Noland added a comment - Should we change it's name to hadoop-2? Also I think 2.0.2-alpha is out: http://www.apache.org/dist/hadoop/common/
      Hide
      Hari Shreedharan added a comment -

      I think leaving the name as such is fine for now. Also, since we do not have any explicit dependencies on hadoop-2.0.2-alpha, leaving it at an older version is probably ok. Upgrading to a newer version of hadoop is often tricky

      Show
      Hari Shreedharan added a comment - I think leaving the name as such is fine for now. Also, since we do not have any explicit dependencies on hadoop-2.0.2-alpha, leaving it at an older version is probably ok. Upgrading to a newer version of hadoop is often tricky
      Hide
      Roman Shaposhnik added a comment -

      +1 on that

      Show
      Roman Shaposhnik added a comment - +1 on that
      Hide
      Brock Noland added a comment -

      I don't see any reason why going from 0.23.1 to 2.0.0-alpha would less tricky than 2.0.2-alpha but I am not to concerned about that since it's just for building/testing purposes.

      But why wouldn't we update the name? It's not 0.23, it's 2.0?

      Show
      Brock Noland added a comment - I don't see any reason why going from 0.23.1 to 2.0.0-alpha would less tricky than 2.0.2-alpha but I am not to concerned about that since it's just for building/testing purposes. But why wouldn't we update the name? It's not 0.23, it's 2.0?
      Hide
      Roman Shaposhnik added a comment -

      Its both actually. Or neither (and no I'm not being Zen). From where I stand I agree with Hari that leaving things as they are is fine especially since it has an added, albeit small, benefit of aligning nicely with other projects (HBase, Pig, etc all use 23 profile as a name). On the other hand if we want to pick the right name for this profile it should probably be disassociated from versions, since I would imaging both camps (hadoop 2.X and hadoop 0.23) using the same profile. So we should probably call it 'next' or 'ng' or something like that.

      Just my 2c.

      Show
      Roman Shaposhnik added a comment - Its both actually. Or neither (and no I'm not being Zen). From where I stand I agree with Hari that leaving things as they are is fine especially since it has an added, albeit small, benefit of aligning nicely with other projects (HBase, Pig, etc all use 23 profile as a name). On the other hand if we want to pick the right name for this profile it should probably be disassociated from versions, since I would imaging both camps (hadoop 2.X and hadoop 0.23) using the same profile. So we should probably call it 'next' or 'ng' or something like that. Just my 2c.
      Hide
      Brock Noland added a comment -

      It might depend on the branch/tag but both 0.94 and 0.94.2 calls 2.X hadoop-2 and 0.23.X hadoop-0.23 while trunk only mentions hadoop 1, 2, and 3. If it's 2.x, let's just call it hadoop-2.

      Show
      Brock Noland added a comment - It might depend on the branch/tag but both 0.94 and 0.94.2 calls 2.X hadoop-2 and 0.23.X hadoop-0.23 while trunk only mentions hadoop 1, 2, and 3. If it's 2.x, let's just call it hadoop-2.
      Hide
      Hari Shreedharan added a comment -

      I am going to update the profile name to hadoop-2. Since this is not a user facing change(only developers building flume will be impacted - I am going to assume they will look at the pom to figure it out), I am going to assume it is not a major issue.

      Show
      Hari Shreedharan added a comment - I am going to update the profile name to hadoop-2. Since this is not a user facing change(only developers building flume will be impacted - I am going to assume they will look at the pom to figure it out), I am going to assume it is not a major issue.
      Hide
      Hari Shreedharan added a comment -

      Updating to hadoop-2 everywhere.

      Show
      Hari Shreedharan added a comment - Updating to hadoop-2 everywhere.
      Hide
      Brock Noland added a comment -

      +1 I will run tests and commit this assuming no problems.

      Show
      Brock Noland added a comment - +1 I will run tests and commit this assuming no problems.
      Hide
      Brock Noland added a comment - - edited

      Hari, the patch doesn't want to apply:

      $ git pull
      Already up-to-date.
      $ git status
      # On branch trunk
      nothing to commit (working directory clean)
      $ patch --dry-run < /data/patches/FLUME-1653.patch 
      patching file pom.xml
      Hunk #1 succeeded at 88 with fuzz 1 (offset -25 lines).
      patching file pom.xml
      Hunk #1 succeeded at 88 with fuzz 1 (offset 3 lines).
      patching file pom.xml
      Hunk #1 FAILED at 125.
      1 out of 1 hunk FAILED -- saving rejects to file pom.xml.rej
      patching file pom.xml
      Hunk #1 succeeded at 88 with fuzz 2 (offset -40 lines).
      patching file pom.xml
      
      Show
      Brock Noland added a comment - - edited Hari, the patch doesn't want to apply: $ git pull Already up-to-date. $ git status # On branch trunk nothing to commit (working directory clean) $ patch --dry-run < /data/patches/FLUME-1653.patch patching file pom.xml Hunk #1 succeeded at 88 with fuzz 1 (offset -25 lines). patching file pom.xml Hunk #1 succeeded at 88 with fuzz 1 (offset 3 lines). patching file pom.xml Hunk #1 FAILED at 125. 1 out of 1 hunk FAILED -- saving rejects to file pom.xml.rej patching file pom.xml Hunk #1 succeeded at 88 with fuzz 2 (offset -40 lines). patching file pom.xml
      Hide
      Hari Shreedharan added a comment -

      Brock - I think you downloaded the old patch. It is applying cleanly for me. Can you download the current file and try again?

      Haris-MacBook-Pro:flume hshreedharan$ git pull
      Already up-to-date.
      Haris-MacBook-Pro:flume hshreedharan$ git apply ~/work/FLUME-1653.patch
      Haris-MacBook-Pro:flume hshreedharan$ 
      
      Show
      Hari Shreedharan added a comment - Brock - I think you downloaded the old patch. It is applying cleanly for me. Can you download the current file and try again? Haris-MacBook-Pro:flume hshreedharan$ git pull Already up-to-date. Haris-MacBook-Pro:flume hshreedharan$ git apply ~/work/FLUME-1653.patch Haris-MacBook-Pro:flume hshreedharan$
      Hide
      Brock Noland added a comment -

      Patch doesn't apply the patch but git apply does.

      Show
      Brock Noland added a comment - Patch doesn't apply the patch but git apply does.
      Hide
      Brock Noland added a comment -

      Committed to trunk, 1.4, and 1.3. Thanks Hari!

      Show
      Brock Noland added a comment - Committed to trunk, 1.4, and 1.3. Thanks Hari!
      Hide
      Hudson added a comment -

      Integrated in flume-trunk #320 (See https://builds.apache.org/job/flume-trunk/320/)
      FLUME-1653: Update Hadoop-23 profile to point to hadoop-2 alpha artifacts (Revision 831a86fc5501a8624b184ea65e53749df31692b8)

      Result = SUCCESS
      brock : http://git-wip-us.apache.org/repos/asf/flume/repo?p=flume.git&a=commit&h=831a86fc5501a8624b184ea65e53749df31692b8
      Files :

      • flume-ng-sinks/flume-hdfs-sink/pom.xml
      • flume-ng-sinks/flume-ng-hbase-sink/pom.xml
      • flume-ng-channels/flume-recoverable-memory-channel/pom.xml
      • pom.xml
      • flume-ng-channels/flume-file-channel/pom.xml
      Show
      Hudson added a comment - Integrated in flume-trunk #320 (See https://builds.apache.org/job/flume-trunk/320/ ) FLUME-1653 : Update Hadoop-23 profile to point to hadoop-2 alpha artifacts (Revision 831a86fc5501a8624b184ea65e53749df31692b8) Result = SUCCESS brock : http://git-wip-us.apache.org/repos/asf/flume/repo?p=flume.git&a=commit&h=831a86fc5501a8624b184ea65e53749df31692b8 Files : flume-ng-sinks/flume-hdfs-sink/pom.xml flume-ng-sinks/flume-ng-hbase-sink/pom.xml flume-ng-channels/flume-recoverable-memory-channel/pom.xml pom.xml flume-ng-channels/flume-file-channel/pom.xml

        People

        • Assignee:
          Hari Shreedharan
          Reporter:
          Hari Shreedharan
        • Votes:
          0 Vote for this issue
          Watchers:
          5 Start watching this issue

          Dates

          • Created:
            Updated:
            Resolved:

            Development