Partly out of curiosity I am trying to reproduce https://issues.apache.org/jira/browse/ZOOKEEPER-2355 in a 3 node zookeeper ensemble running 3.4.6 or 3.4.8.
I am using a debugger and/or hacking the code to try to follow the steps to reproduce but I am not getting behavior that matches the description. - 1. Configure three node ZooKeeper cluster, lets say nodes are A, B and C, start all, assume A is leader, B and C are follower - 2. Connect to any of the server and create ephemeral node /e1 - 3. Close the session, ephemeral node /e1 will go for deletion - 4. While receiving delete proposal make Follower B to fail with SocketTimeoutException. This we need to do to reproduce the scenario otherwise in production environment it happens because of network fault. - ... On step #4, I do not receive a "delete" proposal, I receive a "close session" proposal. If I throw an exception at that point, I do not witness the "Stuck" znode. Is there supposed to be a "delete" proposal when a session disconnects that has existing ephemerals? Looking more closely at the bug description I do not see that it affects 3.4.6 -- is this true do you think or just an omission? I see the same behavior in 3.4.8 which IS supposedly subject to this bug - ie I cannot reproduce the bug there. Thanks!
