Details

    • Type: Task
    • Status: Resolved
    • Priority: Major
    • Resolution: Fixed
    • Affects Version/s: 3.0.0-alpha1
    • Fix Version/s: 2.8.0, 3.0.0-alpha1
    • Component/s: build
    • Labels:
      None
    • Target Version/s:
    • Hadoop Flags:
      Incompatible change
    • Release Note:
      With the introduction of the markdown-formatted and automatically built changes file, the CHANGES.txt files have been eliminated.

      Description

      With the commit of HADOOP-11731, the CHANGES.txt files are now EOLed. We should remove them.

        Issue Links

          Activity

          Hide
          aw Allen Wittenauer added a comment -
          Show
          aw Allen Wittenauer added a comment - FYI: HADOOP-12892
          Hide
          aw Allen Wittenauer added a comment -

          mvn site due to a bug in doxia's markdown parser will go into an infinite loop on unmatched _'s. yetus 0.1.0 + the release note contents worked around it in such a way that broke github rendering. yetus 0.2.0 + the release note contents are now real markdown such that github rendering works. So that problem will go away as soon as yetus 0.2.0 is released (vote closes in a few days) and hadoop is configured to use that version by default. in the mean time, you can unpack the yetus 0.2.0 RC2 in a dir and point YETUS_HOME to it. that will trigger the rdm in that version instead.

          Show
          aw Allen Wittenauer added a comment - mvn site due to a bug in doxia's markdown parser will go into an infinite loop on unmatched _'s. yetus 0.1.0 + the release note contents worked around it in such a way that broke github rendering. yetus 0.2.0 + the release note contents are now real markdown such that github rendering works. So that problem will go away as soon as yetus 0.2.0 is released (vote closes in a few days) and hadoop is configured to use that version by default. in the mean time, you can unpack the yetus 0.2.0 RC2 in a dir and point YETUS_HOME to it. that will trigger the rdm in that version instead.
          Hide
          andrew.wang Andrew Wang added a comment -

          I suspect it's going to be a lot more than that. Would you rather I take a crack at it?

          That'd be great. Right now my machine is hanging when calling "mvn site -Preleasedocs", which really put a damper on my testing efforts. I was about to write you an email, but this is the last log line with debug logging on:

          [DEBUG] Generating /home/andrew/dev/hadoop/trunk/hadoop-common-project/hadoop-common/target/site/release/3.0.0-SNAPSHOT/RELEASENOTES.3.0.0-SNAPSHOT.html
          
          Show
          andrew.wang Andrew Wang added a comment - I suspect it's going to be a lot more than that. Would you rather I take a crack at it? That'd be great. Right now my machine is hanging when calling "mvn site -Preleasedocs", which really put a damper on my testing efforts. I was about to write you an email, but this is the last log line with debug logging on: [DEBUG] Generating /home/andrew/dev/hadoop/trunk/hadoop-common-project/hadoop-common/target/site/release/3.0.0-SNAPSHOT/RELEASENOTES.3.0.0-SNAPSHOT.html
          Hide
          aw Allen Wittenauer added a comment -

          I'm going to fix the release script, I think it's basically just adding "-Preleasedocs"? Will do some before-and-afters to compare.

          I suspect it's going to be a lot more than that. Would you rather I take a crack at it?

          Cleaning up the releasedocmaker lint errors will be a labor of love. I did some 3.0.0 cleanup earlier this week, but did not tackle the branch-2 releases. Ultimately this falls on the 2.8/2.9 RMs, but anyone with JIRA permissions can help out with this.

          I did a big pass through the 5 years of 3.x issues and all of the 2.x issues late last year, with some touch up a month or so ago. They should be in pretty good shape really.

          Show
          aw Allen Wittenauer added a comment - I'm going to fix the release script, I think it's basically just adding "-Preleasedocs"? Will do some before-and-afters to compare. I suspect it's going to be a lot more than that. Would you rather I take a crack at it? Cleaning up the releasedocmaker lint errors will be a labor of love. I did some 3.0.0 cleanup earlier this week, but did not tackle the branch-2 releases. Ultimately this falls on the 2.8/2.9 RMs, but anyone with JIRA permissions can help out with this. I did a big pass through the 5 years of 3.x issues and all of the 2.x issues late last year, with some touch up a month or so ago. They should be in pretty good shape really.
          Hide
          andrew.wang Andrew Wang added a comment -

          I'm going to fix the release script, I think it's basically just adding "-Preleasedocs"? Will do some before-and-afters to compare.

          Cleaning up the releasedocmaker lint errors will be a labor of love. I did some 3.0.0 cleanup earlier this week, but did not tackle the branch-2 releases. Ultimately this falls on the 2.8/2.9 RMs, but anyone with JIRA permissions can help out with this.

          Show
          andrew.wang Andrew Wang added a comment - I'm going to fix the release script, I think it's basically just adding "-Preleasedocs"? Will do some before-and-afters to compare. Cleaning up the releasedocmaker lint errors will be a labor of love. I did some 3.0.0 cleanup earlier this week, but did not tackle the branch-2 releases. Ultimately this falls on the 2.8/2.9 RMs, but anyone with JIRA permissions can help out with this.
          Hide
          aw Allen Wittenauer added a comment -

          rdm's lint mode also tags multiple versions as broken. Unfortunately, Hadoop has a ton of committers and PMCs that really have no idea how to properly close things in JIRA.

          Show
          aw Allen Wittenauer added a comment - rdm's lint mode also tags multiple versions as broken. Unfortunately, Hadoop has a ton of committers and PMCs that really have no idea how to properly close things in JIRA.
          Hide
          mzuehlke Marco Zühlke added a comment -

          With this change we are relying on the correctness of the JIRA fixVersion field.I think this is the right move.

          The wiki has the following guide:

          Always set the "Fix Version" at this point, but please only set a single fix version, the earliest release in which the change will appear. Special case- when committing to a non-mainline branch (such as branch-0.22 or branch-0.23 ATM), please set fix-version to either 2.x.x or 3.x.x appropriately too.

          I my eyes this would (currently) for most issues mean:

          • exactly one of "3.0.0", "2.9.0" or "2.8.0"
          • in addition for backports to branch-2.7 add "2.7.3"
          • in addition for backports to branch-2.6 add "2.6.5"

          Exceptions are issues are are only relevant for a specific branch.

          But when browsing over all fixed JIRAs you see wide variety:

          https://issues.apache.org/jira/issues/?jql=project%20in%20%28YARN%2C%20HADOOP%2C%20HDFS%2C%20MAPREDUCE%29%20AND%20%20%20fixVersion%20in%283.0.0%2C2.9.0%2C2.8.0%2C2.7.0%2C2.7.1%2C2.7.2%2C2.7.3%2C2.6.0%2C2.6.1%2C2.6.2%2C2.6.3%2C2.6.4%2C2.6.5%29%20%20ORDER%20BY%20updated%20DESC

          To just pick some:

          Show
          mzuehlke Marco Zühlke added a comment - With this change we are relying on the correctness of the JIRA fixVersion field.I think this is the right move. The wiki has the following guide: Always set the "Fix Version" at this point, but please only set a single fix version, the earliest release in which the change will appear. Special case- when committing to a non-mainline branch (such as branch-0.22 or branch-0.23 ATM), please set fix-version to either 2.x.x or 3.x.x appropriately too. I my eyes this would (currently) for most issues mean: exactly one of "3.0.0", "2.9.0" or "2.8.0" in addition for backports to branch-2.7 add "2.7.3" in addition for backports to branch-2.6 add "2.6.5" Exceptions are issues are are only relevant for a specific branch. But when browsing over all fixed JIRAs you see wide variety: https://issues.apache.org/jira/issues/?jql=project%20in%20%28YARN%2C%20HADOOP%2C%20HDFS%2C%20MAPREDUCE%29%20AND%20%20%20fixVersion%20in%283.0.0%2C2.9.0%2C2.8.0%2C2.7.0%2C2.7.1%2C2.7.2%2C2.7.3%2C2.6.0%2C2.6.1%2C2.6.2%2C2.6.3%2C2.6.4%2C2.6.5%29%20%20ORDER%20BY%20updated%20DESC To just pick some: YARN-4344 : 2.7.2, 2.6.3 HDFS-9855 : 2.8.0, 2.9.0 HADOOP-12841 : 3.0.0, 2.9.0 YARN-4722 : 2.8.0, 2.7.3, 2.9.0, 2.6.5 ...
          Hide
          aw Allen Wittenauer added a comment -

          Is someone going to fix the release script now?

          Show
          aw Allen Wittenauer added a comment - Is someone going to fix the release script now?
          Hide
          hudson Hudson added a comment -

          FAILURE: Integrated in Hadoop-trunk-Commit #9422 (See https://builds.apache.org/job/Hadoop-trunk-Commit/9422/)
          HADOOP-11792. Remove all of the CHANGES.txt files. (wang: rev 1bc454615e916e66b132eb688a4bc8e76dcb0820)

          • hadoop-common-project/hadoop-common/CHANGES.txt
          • hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt
          • hadoop-mapreduce-project/CHANGES.txt
          • hadoop-yarn-project/CHANGES.txt
          Show
          hudson Hudson added a comment - FAILURE: Integrated in Hadoop-trunk-Commit #9422 (See https://builds.apache.org/job/Hadoop-trunk-Commit/9422/ ) HADOOP-11792 . Remove all of the CHANGES.txt files. (wang: rev 1bc454615e916e66b132eb688a4bc8e76dcb0820) hadoop-common-project/hadoop-common/CHANGES.txt hadoop-hdfs-project/hadoop-hdfs/CHANGES.txt hadoop-mapreduce-project/CHANGES.txt hadoop-yarn-project/CHANGES.txt
          Hide
          andrew.wang Andrew Wang added a comment -

          Since we just resolved HADOOP-12651, I blew away CHANGES.txt on trunk, branch-2, and branch-2.8. Will send a notice to common-dev, happy committing all!

          Show
          andrew.wang Andrew Wang added a comment - Since we just resolved HADOOP-12651 , I blew away CHANGES.txt on trunk, branch-2, and branch-2.8. Will send a notice to common-dev, happy committing all!
          Hide
          andrew.wang Andrew Wang added a comment -

          Assigning to myself for tracking. My current plan is to bring HADOOP-12651 back to branch-2 before going forward with this.

          Show
          andrew.wang Andrew Wang added a comment - Assigning to myself for tracking. My current plan is to bring HADOOP-12651 back to branch-2 before going forward with this.
          Hide
          ajisakaa Akira Ajisaka added a comment -

          +1 for removing CHANGES.txt from trunk, branch-2, and branch-2.8.

          Show
          ajisakaa Akira Ajisaka added a comment - +1 for removing CHANGES.txt from trunk, branch-2, and branch-2.8.
          Hide
          andrew.wang Andrew Wang added a comment -

          Based on the positive mailing list discussion, I'm just going to remove CHANGES.txt from trunk and branch-2 unless there are other objections. I don't plan to post a patch, since CHANGES.txt changes have never gone through Jenkins or required a +1. It'll literally be running git rm **/CHANGES.txt and committing it referencing this JIRA.

          Show
          andrew.wang Andrew Wang added a comment - Based on the positive mailing list discussion, I'm just going to remove CHANGES.txt from trunk and branch-2 unless there are other objections. I don't plan to post a patch, since CHANGES.txt changes have never gone through Jenkins or required a +1. It'll literally be running git rm **/CHANGES.txt and committing it referencing this JIRA.
          Hide
          aw Allen Wittenauer added a comment -

          This is only targeting trunk. branch-2 is too old, too crufty, and too fundamentally broken for my time to be wasted on it.

          Show
          aw Allen Wittenauer added a comment - This is only targeting trunk. branch-2 is too old, too crufty, and too fundamentally broken for my time to be wasted on it.
          Hide
          szetszwo Tsz Wo Nicholas Sze added a comment -

          I think we should not remove CHANGES.txt until the next release; see also this comment.

          Show
          szetszwo Tsz Wo Nicholas Sze added a comment - I think we should not remove CHANGES.txt until the next release; see also this comment .

            People

            • Assignee:
              andrew.wang Andrew Wang
              Reporter:
              aw Allen Wittenauer
            • Votes:
              2 Vote for this issue
              Watchers:
              15 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development