flink's hard dependency on zookeeper for HA

[Vishwas Siravara <vs...@gmail.com>]

Hi all,
I am using flink 1.7.2 as a standalone cluster in high availability mode
with zookeeper. I have noticed that all flink processes go down once
zookeeper goes down ? Is this expected behavior since the leader election
has already happened and the job has been running for several hours.


Best,
Vishwas

Re: flink's hard dependency on zookeeper for HA

[vino yang <ya...@gmail.com>]

Hi Vishwas,

In the standalone cluster HA mode, Flink heavily depends on ZooKeeper. Not
only for leader election, but also for:


   - Checkpoint metadata info;
   - JobGraph store;
   - ....

So you should make sure your ZooKeeper Cluster works normally. More details
please see[1][2].

Best,
Vino

[1]:
https://cwiki.apache.org/confluence/display/FLINK/JobManager+High+Availability
[2]:
https://ci.apache.org/projects/flink/flink-docs-master/ops/jobmanager_high_availability.html

Vishwas Siravara <vs...@gmail.com> 于2019年11月7日周四 上午12:07写道：

> Hi all,
> I am using flink 1.7.2 as a standalone cluster in high availability mode
> with zookeeper. I have noticed that all flink processes go down once
> zookeeper goes down ? Is this expected behavior since the leader election
> has already happened and the job has been running for several hours.
>
>
> Best,
> Vishwas
>