> On March 4, 2015, 1:59 a.m., Jun Rao wrote: > > core/src/main/scala/kafka/utils/ZkUtils.scala, lines 817-826 > > <https://reviews.apache.org/r/31711/diff/1/?file=884018#file884018line817> > > > > We will need to remember the result of client.exist(). If > > isNamespaceChecked is true, we will throw a ConfigException if the > > namespace doesn't exist.
Not sure if that is a good idea. If we do that then there won't be an easy way to fix the situation. Even if someone creates that namespace on ZK, ZkPath will keep complaining about the missing namespace. With current logic, if there is an exception then namespace will be rechecked the next time as well. Let me know if this does not make sense. - Ashish ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/31711/#review75132 ----------------------------------------------------------- On March 4, 2015, 12:45 a.m., Ashish Singh wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/31711/ > ----------------------------------------------------------- > > (Updated March 4, 2015, 12:45 a.m.) > > > Review request for kafka. > > > Bugs: KAFKA-1994 > https://issues.apache.org/jira/browse/KAFKA-1994 > > > Repository: kafka > > > Description > ------- > > KAFKA-1994: Evaluate performance effect of chroot check on Topic creation > > > Diffs > ----- > > core/src/main/scala/kafka/utils/ZkUtils.scala > 7ae999ec619443d35a9cb8fbcd531fca0c51c8c0 > core/src/test/scala/unit/kafka/zk/ZKPathTest.scala > 9897b2fa8f8261fe8ab6b62b45b9052adb07043f > > Diff: https://reviews.apache.org/r/31711/diff/ > > > Testing > ------- > > > Thanks, > > Ashish Singh > >
