Here's two test cases reproducing the error.

Both use pykeeper (ease_install pykeeper) to set things up, but then use the zkpython bindings to do the actual testing.

zktest3.py leaks, zktest4.py does not.

https://github.com/apache/zookeeper/blob/trunk/src/contrib/zkpython/src/c/zookeeper.c#L487

Shouldn't arglist be decref'ed after the call?

That drops my tuple count from ~55k to ~5k after a zktest3.py run.

This bug applies to ALL async api calls with python.

Fix for async python api memory leaks.

Noting the problem is generic, it also applies to all extant versions 3.3x and 3.4

Also wanted to note fwiw, we're going to apply the patch i attached to this issue, in the zookeeper 3.3.5 pkg distributed in ubuntu 12.04/precise.

-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12519666/pyzk-mem-leak-fix.diff
against trunk revision 1330043.

+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 new tests are needed for this patch.
Also please list what manual steps were performed to verify this patch.

-1 patch. The patch command could not apply the patch.

Console output: https://builds.apache.org/job/PreCommit-ZOOKEEPER-Build/1050//console

This message is automatically generated.

Kapil the patch is failing to apply. Please update it and re-attach, thanks. (you might check if the patch applies cleanly to both 3.4 and trunk branches, it might be that it only applies to one, in which case you would need to submit two patches, one for each branch). thanks!

Corrected patch that applies to trunk and 3.4 branch.

-1 overall. Here are the results of testing the latest attachment
http://issues.apache.org/jira/secure/attachment/12526345/zk.patch
against trunk revision 1336467.

+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 new tests are needed for this patch.
Also please list what manual steps were performed to verify 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 (version 1.3.9) warnings.

+1 release audit. The applied patch does not increase the total number of release audit warnings.

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

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

Test results: https://builds.apache.org/job/PreCommit-ZOOKEEPER-Build/1070//testReport/
Findbugs warnings: https://builds.apache.org/job/PreCommit-ZOOKEEPER-Build/1070//artifact/trunk/build/test/findbugs/newPatchFindbugsWarnings.html
Console output: https://builds.apache.org/job/PreCommit-ZOOKEEPER-Build/1070//console

This message is automatically generated.

Is this patch going to be applied soon? Are you waiting for justification on why it doesn't include new tests?

I would also like to know. Tests were not included because it's not easy to test for a memory leak.

Patch looks good, I'll commit shortly.

Just committed to 3.3, 3.4 and 3.5. Thanks Kapil and Andre!

Integrated in ZooKeeper-trunk #1590 (See https://builds.apache.org/job/ZooKeeper-trunk/1590/)
ZOOKEEPER-1431. zkpython async calls leak memory (Kapil Thangavelu and Andre Cruz via henryr) (Revision 1351541)

Result = SUCCESS
henry : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVN&view=rev&rev=1351541
Files :

• /zookeeper/trunk/CHANGES.txt
• /zookeeper/trunk/src/contrib/zkpython/src/c/zookeeper.c