Hadoop Common
  1. Hadoop Common
  2. HADOOP-4104

add time, permission and user attribute support to libhdfs

    Details

    • Type: New Feature New Feature
    • Status: Closed
    • Priority: Major Major
    • Resolution: Fixed
    • Affects Version/s: None
    • Fix Version/s: 0.19.0
    • Component/s: None
    • Labels:
      None
    • Hadoop Flags:
      Reviewed

      Description

      1. add mtime, accesstime, owner, permissions to HDFSFileInfo
      2. create hdfsChown, hdfsChmod, hdfsUtime functions
      3. create hdfsConnectAsUser function

      1. HADOOP-4104.2.txt
        24 kB
        Pete Wyckoff
      2. HADOOP-4104.3.txt
        26 kB
        Pete Wyckoff
      3. HADOOP-4104.4.txt
        28 kB
        Pete Wyckoff
      4. HADOOP-4104.txt
        36 kB
        Pete Wyckoff

        Issue Links

          Activity

          Hide
          Pete Wyckoff added a comment -

          This is the combination of 3 previous jiras but since all really related, I am combining them and supplying 1 patch.

          Show
          Pete Wyckoff added a comment - This is the combination of 3 previous jiras but since all really related, I am combining them and supplying 1 patch.
          Hide
          Pete Wyckoff added a comment -

          NOTE: this includes https://issues.apache.org/jira/browse/HADOOP-3963 which is waiting to be committed.

          Otherwise, this patch is basically complete - all tests have been added and pass.

          One caveat is setting/getting access times. Since there's no way right now to enable this in the libhdfs testing framework, there's really no way to have a unit test for it.

          Show
          Pete Wyckoff added a comment - NOTE: this includes https://issues.apache.org/jira/browse/HADOOP-3963 which is waiting to be committed. Otherwise, this patch is basically complete - all tests have been added and pass. One caveat is setting/getting access times. Since there's no way right now to enable this in the libhdfs testing framework, there's really no way to have a unit test for it.
          Hide
          Craig Macdonald added a comment -

          Pete,

          Looks good so far.

          • I was wondering if hdfsConnectAsUser should fail to work unless geteuid() == 0
          Show
          Craig Macdonald added a comment - Pete, Looks good so far. I was wondering if hdfsConnectAsUser should fail to work unless geteuid() == 0
          Hide
          Pete Wyckoff added a comment -

          Hi Craig,

          Yes, I agree that's a good idea - probably should be enforced in the filesystem java api? we should get Nicholas to comment on why they haven't done this; now that you bring this up, I'm surprised it doesn't raise an exception - I ran the unit tests as me and not root.

          pete

          Show
          Pete Wyckoff added a comment - Hi Craig, Yes, I agree that's a good idea - probably should be enforced in the filesystem java api? we should get Nicholas to comment on why they haven't done this; now that you bring this up, I'm surprised it doesn't raise an exception - I ran the unit tests as me and not root. pete
          Hide
          Craig Macdonald added a comment -

          My understanding is that Hadoop has no real authentication in it - it just assumes that what UGI gives is correct. This is possibly the first scenario where we're bending UGI to give something different. That being the case, we should provide the security.

          I think any security checks in Hadoop will depend how Hadoop implements authentication and security in due course. In the meantime, it's probably good that we enforce some security in the mean time.

          My understanding from http://en.allexperts.com/q/Unix-Linux-OS-1064/real-effective-user-id.htm is that what we want to check is effective uid. whoami (at least on linux) will return euid, not uid.

          Show
          Craig Macdonald added a comment - My understanding is that Hadoop has no real authentication in it - it just assumes that what UGI gives is correct. This is possibly the first scenario where we're bending UGI to give something different. That being the case, we should provide the security. I think any security checks in Hadoop will depend how Hadoop implements authentication and security in due course. In the meantime, it's probably good that we enforce some security in the mean time. My understanding from http://en.allexperts.com/q/Unix-Linux-OS-1064/real-effective-user-id.htm is that what we want to check is effective uid. whoami (at least on linux) will return euid, not uid.
          Hide
          Hadoop QA added a comment -

          -1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12389646/HADOOP-4104.txt
          against trunk revision 692700.

          +1 @author. The patch does not contain any @author tags.

          -1 tests included. The patch doesn't appear to include any new or modified tests.
          Please justify why no tests are needed for this patch.

          +1 javadoc. The javadoc tool did not generate any warning messages.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 findbugs. The patch does not introduce any new Findbugs warnings.

          -1 core tests. The patch failed core unit tests.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/testReport/
          Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
          Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/artifact/trunk/build/test/checkstyle-errors.html
          Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - -1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12389646/HADOOP-4104.txt against trunk revision 692700. +1 @author. The patch does not contain any @author tags. -1 tests included. The patch doesn't appear to include any new or modified tests. Please justify why no tests are needed for this patch. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. -1 core tests. The patch failed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/artifact/trunk/build/test/checkstyle-errors.html Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3200/console This message is automatically generated.
          Hide
          Hadoop QA added a comment -

          -1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12389646/HADOOP-4104.txt
          against trunk revision 692700.

          +1 @author. The patch does not contain any @author tags.

          -1 tests included. The patch doesn't appear to include any new or modified tests.
          Please justify why no tests are needed for this patch.

          +1 javadoc. The javadoc tool did not generate any warning messages.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 findbugs. The patch does not introduce any new Findbugs warnings.

          +1 core tests. The patch passed core unit tests.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/testReport/
          Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
          Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/artifact/trunk/build/test/checkstyle-errors.html
          Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - -1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12389646/HADOOP-4104.txt against trunk revision 692700. +1 @author. The patch does not contain any @author tags. -1 tests included. The patch doesn't appear to include any new or modified tests. Please justify why no tests are needed for this patch. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. +1 core tests. The patch passed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/artifact/trunk/build/test/checkstyle-errors.html Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3201/console This message is automatically generated.
          Hide
          Pete Wyckoff added a comment -

          in that case, shouldn't the dfsclient enforce it too?

          RE: -1 unit tests – Craig and I added a bunch of them to hdfs_test.c just hudson doesn't know!

          Show
          Pete Wyckoff added a comment - in that case, shouldn't the dfsclient enforce it too? RE: -1 unit tests – Craig and I added a bunch of them to hdfs_test.c just hudson doesn't know!
          Hide
          dhruba borthakur added a comment -

          My opinion would be to avoid adding security and access checks that are specific to libhdfs only. Does libhdfs work on non-unix platforms? If so, then we should avoid using calls like geteuid(). It will keep the library very-hadoop-centric rather than being unix-centric.

          Show
          dhruba borthakur added a comment - My opinion would be to avoid adding security and access checks that are specific to libhdfs only. Does libhdfs work on non-unix platforms? If so, then we should avoid using calls like geteuid(). It will keep the library very-hadoop-centric rather than being unix-centric.
          Hide
          Sameer Paranjpye added a comment -

          My opinion would be to avoid adding security and access checks that are specific to libhdfs only. Does libhdfs work on non-unix platforms? If so, then we should avoid using calls like geteuid(). It will keep the library very-hadoop-centric rather than being unix-centric.

          +1

          Hadoop doesn't have authentication. The ugi and permissions implementation is aimed at preventing mistakes rather than providing any real security. I don't see why libhdfs should take it upon itself to do more. Any future authentication work in Hadoop can be leveraged by libhdfs.

          Show
          Sameer Paranjpye added a comment - My opinion would be to avoid adding security and access checks that are specific to libhdfs only. Does libhdfs work on non-unix platforms? If so, then we should avoid using calls like geteuid(). It will keep the library very-hadoop-centric rather than being unix-centric. +1 Hadoop doesn't have authentication. The ugi and permissions implementation is aimed at preventing mistakes rather than providing any real security. I don't see why libhdfs should take it upon itself to do more. Any future authentication work in Hadoop can be leveraged by libhdfs.
          Hide
          Pete Wyckoff added a comment -

          This is the patch without the hadoop-4113 diff since 4113 has been committed.

          Show
          Pete Wyckoff added a comment - This is the patch without the hadoop-4113 diff since 4113 has been committed.
          Hide
          Pete Wyckoff added a comment -

          submitting final patch for QA

          Show
          Pete Wyckoff added a comment - submitting final patch for QA
          Hide
          Pete Wyckoff added a comment -

          Sameer and Dhrubas have a good point. I think we should then leave the patch as is. Craig - what do you think? The only thing I removed from the patch were the 4113 changes since they are already committed. Otherwise, I haven't changed anything from the patch you looked at.

          Show
          Pete Wyckoff added a comment - Sameer and Dhrubas have a good point. I think we should then leave the patch as is. Craig - what do you think? The only thing I removed from the patch were the 4113 changes since they are already committed. Otherwise, I haven't changed anything from the patch you looked at.
          Hide
          Pete Wyckoff added a comment -

          should make one other comment since 1/2 of this code is from Craig himself in hadoop-3264. That part of the code I have reviewed before I inserted it and it looks good and I think may have also been looked at by Nicolas.

          – pete

          Show
          Pete Wyckoff added a comment - should make one other comment since 1/2 of this code is from Craig himself in hadoop-3264. That part of the code I have reviewed before I inserted it and it looks good and I think may have also been looked at by Nicolas. – pete
          Hide
          Craig Macdonald added a comment -

          Yes, I agree. Hadoop doesnt have a authentication system, so we shouldn't force one on it in this patch.

          +1 on HADOOP-4104.2.txt: I have read through the code and am happy with the patch.

          @ Sameer - I agree that libhdfs can leverage any additional authentication functionality. I should just make clear that a requirement of any authentication change should be "impersonation" by a process running with superuser privileges, as provided hdfsConnectAsUser().

          Show
          Craig Macdonald added a comment - Yes, I agree. Hadoop doesnt have a authentication system, so we shouldn't force one on it in this patch. +1 on HADOOP-4104 .2.txt: I have read through the code and am happy with the patch. @ Sameer - I agree that libhdfs can leverage any additional authentication functionality. I should just make clear that a requirement of any authentication change should be "impersonation" by a process running with superuser privileges, as provided hdfsConnectAsUser().
          Hide
          Hadoop QA added a comment -

          -1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12389775/HADOOP-4104.2.txt
          against trunk revision 693587.

          +1 @author. The patch does not contain any @author tags.

          -1 tests included. The patch doesn't appear to include any new or modified tests.
          Please justify why no tests are needed for this patch.

          +1 javadoc. The javadoc tool did not generate any warning messages.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 findbugs. The patch does not introduce any new Findbugs warnings.

          +1 core tests. The patch passed core unit tests.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/testReport/
          Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
          Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/artifact/trunk/build/test/checkstyle-errors.html
          Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - -1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12389775/HADOOP-4104.2.txt against trunk revision 693587. +1 @author. The patch does not contain any @author tags. -1 tests included. The patch doesn't appear to include any new or modified tests. Please justify why no tests are needed for this patch. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. +1 core tests. The patch passed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/artifact/trunk/build/test/checkstyle-errors.html Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3223/console This message is automatically generated.
          Hide
          Pete Wyckoff added a comment -

          RE: -1 Tests - the patch includes a whole bunch of new tests for all the new functionality - its' in hdfs_test.c so hudson doesn't see it.

          Show
          Pete Wyckoff added a comment - RE: -1 Tests - the patch includes a whole bunch of new tests for all the new functionality - its' in hdfs_test.c so hudson doesn't see it.
          Hide
          Pete Wyckoff added a comment -

          noticed a bug. This is not handling conversion to milliseconds in hdfsUtimes. Its params are (should be) in seconds and the hdfs API is in milliseconds so need to convert.

          Show
          Pete Wyckoff added a comment - noticed a bug. This is not handling conversion to milliseconds in hdfsUtimes. Its params are (should be) in seconds and the hdfs API is in milliseconds so need to convert.
          Hide
          Pete Wyckoff added a comment -

          I also fixed the tests to properly check chmod and permissions which it wasn't doing as hdfs only does those things on files, not directories.

          Show
          Pete Wyckoff added a comment - I also fixed the tests to properly check chmod and permissions which it wasn't doing as hdfs only does those things on files, not directories.
          Hide
          Pete Wyckoff added a comment -

          had a bug in handling of times in writes were expecting milliseconds when the API really calls for seconds since a tTime is a time_t which is typically seconds.

          Show
          Pete Wyckoff added a comment - had a bug in handling of times in writes were expecting milliseconds when the API really calls for seconds since a tTime is a time_t which is typically seconds.
          Hide
          Tsz Wo Nicholas Sze added a comment -

          Some comments on HADOOP-4104.3.txt although I don't have much experience on libhdfs:

          • Try referring to the base class UserGroupInformation instead of UnixUserGroupInformation if possible since UnixUserGroupInformation will be replaced in the future.
          • In hdfsChown(...), if owner == NULL, we should not set owner = "root". owner == NULL means owner remains unchanged.
          Show
          Tsz Wo Nicholas Sze added a comment - Some comments on HADOOP-4104 .3.txt although I don't have much experience on libhdfs: Try referring to the base class UserGroupInformation instead of UnixUserGroupInformation if possible since UnixUserGroupInformation will be replaced in the future. In hdfsChown(...), if owner == NULL, we should not set owner = "root". owner == NULL means owner remains unchanged.
          Hide
          Hadoop QA added a comment -

          +1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12389865/HADOOP-4104.3.txt
          against trunk revision 693705.

          +1 @author. The patch does not contain any @author tags.

          +1 tests included. The patch appears to include 2 new or modified tests.

          +1 javadoc. The javadoc tool did not generate any warning messages.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 findbugs. The patch does not introduce any new Findbugs warnings.

          +1 core tests. The patch passed core unit tests.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/testReport/
          Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
          Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/artifact/trunk/build/test/checkstyle-errors.html
          Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - +1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12389865/HADOOP-4104.3.txt against trunk revision 693705. +1 @author. The patch does not contain any @author tags. +1 tests included. The patch appears to include 2 new or modified tests. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. +1 core tests. The patch passed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/artifact/trunk/build/test/checkstyle-errors.html Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3237/console This message is automatically generated.
          Hide
          Pete Wyckoff added a comment -

          cancel to address nicholas' comments.

          Show
          Pete Wyckoff added a comment - cancel to address nicholas' comments.
          Hide
          Pete Wyckoff added a comment -

          UserGroupInformation doesn't have a constructor?? How do I construct it without already having the login information in a Configuration object?

          Show
          Pete Wyckoff added a comment - UserGroupInformation doesn't have a constructor?? How do I construct it without already having the login information in a Configuration object?
          Hide
          Tsz Wo Nicholas Sze added a comment -

          You are right. UserGroupInformation is abstract, so you must create a subclass object. However, the subclass object can be referred as UserGroupInformation to minimize the reference of the subclass. e.g. UserGroupInformation ugi = new UnixUserGroupInformation(...).

          Show
          Tsz Wo Nicholas Sze added a comment - You are right. UserGroupInformation is abstract, so you must create a subclass object. However, the subclass object can be referred as UserGroupInformation to minimize the reference of the subclass. e.g. UserGroupInformation ugi = new UnixUserGroupInformation(...).
          Hide
          Pete Wyckoff added a comment -

          The problem is though that UGI does not support saveToConf, which is what I really want to get at and even that in UUGI, takes a UUGI.

          Is there another way to connect as a specific user given that I know the user and groups?

          Show
          Pete Wyckoff added a comment - The problem is though that UGI does not support saveToConf, which is what I really want to get at and even that in UUGI, takes a UUGI. Is there another way to connect as a specific user given that I know the user and groups?
          Hide
          Tsz Wo Nicholas Sze added a comment -

          It can be done by UGI.setCurrentUGI(...).

          Show
          Tsz Wo Nicholas Sze added a comment - It can be done by UGI.setCurrentUGI(...).
          Hide
          Pete Wyckoff added a comment -

          It can be done by UGI.setCurrentUGI(...).

          But that isn't associated with a specific filesystem connection, so what are the semantics if you have multiple connections?

          Show
          Pete Wyckoff added a comment - It can be done by UGI.setCurrentUGI(...). But that isn't associated with a specific filesystem connection, so what are the semantics if you have multiple connections?
          Hide
          Pete Wyckoff added a comment -

          This patch fixes the bug of setting user to root and group to users when null. I also added some #ifdef-ed out code for using the setCurrentUGI method in UuserGroupInformation but not clear on the semantics wrt multiple connections and also doesn't work.

          This patch passes ant -Dlibhdfs=1 test-libhdfs but will wait and see if there's an easy way to get away from the UnixUserGroupInformation object before submitting it as a patch.

          Show
          Pete Wyckoff added a comment - This patch fixes the bug of setting user to root and group to users when null. I also added some #ifdef-ed out code for using the setCurrentUGI method in UuserGroupInformation but not clear on the semantics wrt multiple connections and also doesn't work. This patch passes ant -Dlibhdfs=1 test-libhdfs but will wait and see if there's an easy way to get away from the UnixUserGroupInformation object before submitting it as a patch.
          Hide
          Pete Wyckoff added a comment -

          I am going to submit as is and once we figure out the UUGI versus UGI since we have a week before 0.19. Can do a quick fix later when UUGI goes away or sometime later.

          Show
          Pete Wyckoff added a comment - I am going to submit as is and once we figure out the UUGI versus UGI since we have a week before 0.19. Can do a quick fix later when UUGI goes away or sometime later.
          Hide
          Tsz Wo Nicholas Sze added a comment -

          > But that isn't associated with a specific filesystem connection, so what are the semantics if you have multiple connections?

          In this case, it seems better using UUGI and conf.

          The patch looks good to me but, again, I don't know much about libhdfs.

          Show
          Tsz Wo Nicholas Sze added a comment - > But that isn't associated with a specific filesystem connection, so what are the semantics if you have multiple connections? In this case, it seems better using UUGI and conf. The patch looks good to me but, again, I don't know much about libhdfs.
          Hide
          Hadoop QA added a comment -

          +1 overall. Here are the results of testing the latest attachment
          http://issues.apache.org/jira/secure/attachment/12389977/HADOOP-4104.4.txt
          against trunk revision 695690.

          +1 @author. The patch does not contain any @author tags.

          +1 tests included. The patch appears to include 2 new or modified tests.

          +1 javadoc. The javadoc tool did not generate any warning messages.

          +1 javac. The applied patch does not increase the total number of javac compiler warnings.

          +1 findbugs. The patch does not introduce any new Findbugs warnings.

          +1 core tests. The patch passed core unit tests.

          +1 contrib tests. The patch passed contrib unit tests.

          Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/testReport/
          Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
          Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/artifact/trunk/build/test/checkstyle-errors.html
          Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/console

          This message is automatically generated.

          Show
          Hadoop QA added a comment - +1 overall. Here are the results of testing the latest attachment http://issues.apache.org/jira/secure/attachment/12389977/HADOOP-4104.4.txt against trunk revision 695690. +1 @author. The patch does not contain any @author tags. +1 tests included. The patch appears to include 2 new or modified tests. +1 javadoc. The javadoc tool did not generate any warning messages. +1 javac. The applied patch does not increase the total number of javac compiler warnings. +1 findbugs. The patch does not introduce any new Findbugs warnings. +1 core tests. The patch passed core unit tests. +1 contrib tests. The patch passed contrib unit tests. Test results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/testReport/ Findbugs warnings: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html Checkstyle results: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/artifact/trunk/build/test/checkstyle-errors.html Console output: http://hudson.zones.apache.org/hudson/job/Hadoop-Patch/3266/console This message is automatically generated.
          Hide
          Pete Wyckoff added a comment -

          Craig, can you peruse the final patch and give a thumbs up?

          Show
          Pete Wyckoff added a comment - Craig, can you peruse the final patch and give a thumbs up?
          Hide
          Craig Macdonald added a comment -

          +1. I've read through, all code looks good. Good to see API docs in libhdfs.h were updated also

          C

          Show
          Craig Macdonald added a comment - +1. I've read through, all code looks good. Good to see API docs in libhdfs.h were updated also C
          Hide
          Zheng Shao added a comment -

          Just committed this. Thanks Pete!
          Committed revision 696741.

          Show
          Zheng Shao added a comment - Just committed this. Thanks Pete! Committed revision 696741.
          Hide
          Arun C Murthy added a comment -

          Resolving issue now that this has been committed.

          Show
          Arun C Murthy added a comment - Resolving issue now that this has been committed.
          Hide
          Hudson added a comment -
          Show
          Hudson added a comment - Integrated in Hadoop-trunk #611 (See http://hudson.zones.apache.org/hudson/job/Hadoop-trunk/611/ )

            People

            • Assignee:
              Pete Wyckoff
              Reporter:
              Pete Wyckoff
            • Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development