[
https://issues.apache.org/jira/browse/FLINK-10052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16886663#comment-16886663
]
lamber-ken commented on FLINK-10052:
------------------------------------
[~Tison], right, it's a better way to upgrate curator dependcy to fix this
ideally, but there's a problem that curator-4.x detect the version of zookeeper
by test whether +org.apache.zookeeper.admin.ZooKeeperAdmin+ is in classpath or
not, like bellow.
{code:java}
Class.forName("org.apache.admin.ZooKeeperAdmin");
{code}
But flink-runtime module shades +org.apache.zookeeper+ to
+org.apache.flink.shaded.zookeeper.org.apache.zookeeper+ , so it'll detect
failed.
I think two ways to fix this issue,
First, rewrite +LeaderLatch#handleStateChange+ at flink-shaded-curator
moduleflink, like [PR#9066|https://github.com/apache/flink/pull/9066].
Seconde, it also could be achievable by using Curator's LeaderSelector instead
of the LeaderLatch as mentioned in issue description
> Tolerate temporarily suspended ZooKeeper connections
> ----------------------------------------------------
>
> Key: FLINK-10052
> URL: https://issues.apache.org/jira/browse/FLINK-10052
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Affects Versions: 1.4.2, 1.5.2, 1.6.0, 1.8.1
> Reporter: Till Rohrmann
> Assignee: Dominik Wosiński
> Priority: Major
>
> This issue results from FLINK-10011 which uncovered a problem with Flink's HA
> recovery and proposed the following solution to harden Flink:
> The {{ZooKeeperLeaderElectionService}} uses the {{LeaderLatch}} Curator
> recipe for leader election. The leader latch revokes leadership in case of a
> suspended ZooKeeper connection. This can be premature in case that the system
> can reconnect to ZooKeeper before its session expires. The effect of the lost
> leadership is that all jobs will be canceled and directly restarted after
> regaining the leadership.
> Instead of directly revoking the leadership upon a SUSPENDED ZooKeeper
> connection, it would be better to wait until the ZooKeeper connection is
> LOST. That way we would allow the system to reconnect and not lose the
> leadership. This could be achievable by using Curator's {{LeaderSelector}}
> instead of the {{LeaderLatch}}.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
