----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/14483/#review26689 -----------------------------------------------------------
Ship it! Thanks Jason, lets make sure we get a integration test where we simulate the disconnects externally. - Kishore Gopalakrishna On Oct. 4, 2013, 5:37 a.m., Zhen Zhang wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/14483/ > ----------------------------------------------------------- > > (Updated Oct. 4, 2013, 5:37 a.m.) > > > Review request for helix, Kanak Biscuitwala and Kishore Gopalakrishna. > > > Repository: helix-git > > > Description > ------- > > If we close zkclient inside its own event thread context, we will get a > ZkInterruptedException and ZkConnection#close() will be skipped. The > workaround catches this ZkInterruptedException and does the > ZkConnection#close(). One tricky point is inside ZkInterruptedException > constructor, current thread is being interrupted again, so we need to clear > interrupted status of current state and safely do the ZkConnection#close. > After that, we need to restore the interrupted status of current thread. > > > Diffs > ----- > > helix-core/src/main/java/org/apache/helix/manager/zk/ZkClient.java 5b3af6d > helix-core/src/test/java/org/apache/helix/ZkTestHelper.java db2a6d0 > helix-core/src/test/java/org/apache/helix/manager/zk/TestZkFlapping.java > e69de29 > > Diff: https://reviews.apache.org/r/14483/diff/ > > > Testing > ------- > > TestZkFlapping > > > Thanks, > > Zhen Zhang > >
