[jira] [Created] (ZOOKEEPER-3466) ZK cluster converges, but does not properly handle client connections (new in 3.5.5)

["Jan-Philip Gehrcke (JIRA)" <ji...@apache.org>]

Jan-Philip Gehrcke created ZOOKEEPER-3466:
---------------------------------------------

             Summary: ZK cluster converges, but does not properly handle client connections (new in 3.5.5)
                 Key: ZOOKEEPER-3466
                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-3466
             Project: ZooKeeper
          Issue Type: Bug
    Affects Versions: 3.5.5
         Environment: Linux
            Reporter: Jan-Philip Gehrcke


Hey, we explore switching from ZooKeeper 3.4.14 to ZooKeeper 3.5.5 in [DC/OS|[https://github.com/dcos/dcos]].

DC/OS coordinates ZooKeeper via Exhibitor. We are not changing anything w.r.t. Exhibitor for now, and are hoping that we can use ZooKeeper 3.5.5 as a drop-in replacement for 3.4.14. This seems to work fine when Exhibitor uses a so-called static ensemble where the individual ZooKeeper instances are known a priori.

When Exhibitor however discovers individual ZooKeeper instances ("dynamic" back-end) then I think we observe a regression where ZooKeeper 3.5.5 can get into the following bad state (often, but not always):
 # three ZooKeeper instances find each other, leader election takes place (*expected*)
 # leader election succeeds: two followers, one leader (*expected*)
 # all three ZK instances respond IAMOK to RUOK  (*expected*)
 # all three ZK instances respond to SRVR (one says "Mode: leader", the other two say "Mode: follower")  (*expected*)
 # all three ZK instances respond to MNTR and show plausible output (*expected*)
 # *{color:#FF0000}Unexpected:{color}* any ZooKeeper client trying to connect to any of the three nodes observes a "connection timeout", whereas notably this is *not* a TCP connect() timeout. The TCP connect() succeeds, but then ZK does not seem to send the bytes to the TCP connection, and the ZK clients wait for them via recv() until it hits a timeout condition. Examples for two different clients:
 ## In Kazoo we specifically hit
Connection time-out: socket time-out during readgenerated here: [https://github.com/python-zk/kazoo/blob/88b657a0977161f3815657878ba48f82a97a3846/kazoo/protocol/connection.py#L249]
 ## In zkCli we see 
Client session timed out, have not heard from server in 15003ms for sessionid 0x0, closing socket connection and attempting reconnect (org.apache.zookeeper.ClientCnxn:main-SendThread(localhost:2181))
 # This state is stable, it will last forever (well, at least for multiple hours and we didn't test longer than that).
 # In our system the ZooKeeper clients are crash-looping. They retry. What I have observed is that while they retry the ZK ensemble accumulates outstanding requests, here shown from MNTR output: 
zk_packets_received	2008
zk_packets_sent	127
zk_num_alive_connections	18
zk_outstanding_requests	1880
 # The leader emits log lines confirming session timeout, example:

{code:java}
[myid:3] INFO [SessionTracker:ZooKeeperServer@398] - Expiring session 0x2000642b18f0020, timeout of 10000ms exceeded
[myid:3] INFO [SessionTracker:QuorumZooKeeperServer@157] - Submitting global closeSession request for session 0x2000642b18f0020{code}

 # In this state, restarting any one of the two ZK followers results in the same state (clients don't get data from ZK upon connect).
 # In this state, restarting the ZK leader, and therefore triggering a leader re-election, almost immediately results in all clients being able to connect to all ZK instances successfully.



--
This message was sent by Atlassian JIRA
(v7.6.14#76016)