The JobManager is the coordinator of each Flink deployment. It is responsible for both scheduling and resource management.
By default, there is a single JobManager instance per Flink cluster. This creates a single point of failure (SPOF): if the JobManager crashes, no new programs can be submitted and running programs fail.
With JobManager High Availability, you can run multiple JobManager instances per Flink cluster and thereby circumvent the SPOF.
The general idea of JobManager high availability is that there is a single leading JobManager at any time and multiple standby JobManagers to take over leadership in case the leader fails. This guarantees that there is no single point of failure and programs can make progress as soon as a standby JobManager has taken leadership. There is no explicit distinction between standby and master JobManager instances. Each JobManager can take the role of master or standby.
As an example, consider the following setup with three JobManager instances:
To enable JobManager High Availability you have to configure a ZooKeeper quorum and set up a masters file with all JobManagers hosts.
Flink leverages ZooKeeper for distributed coordination between all running JobManager instances. ZooKeeper is a separate service from Flink, which provides highly reliable distirbuted coordination via leader election and light-weight consistent state storage. Check out ZooKeeper's Getting Started Guide for more information about ZooKeeper.
Configuring a ZooKeeper quorum in conf/flink-conf.yaml enables high availability mode and all Flink components try to connect to a JobManager via coordination through ZooKeeper.
ZooKeeper quorum (required): A ZooKeeper quorum is a replicated group of ZooKeeper servers, which provide the distributed coordination service.
Each addressX:port refers to a ZooKeeper server, which is reachable by Flink at the given address and port.
The following configuration keys are optional:
ha.zookeeper.dir: /flink [default]: ZooKeeper directory to use for coordinationIn order to start an HA-cluster configure the masters file in conf/masters:
masters file: The masters file contains all hosts, on which JobManagers are started.
jobManagerAddress1 [...] jobManagerAddressX
After configuring the masters and the ZooKeeper quorum, you can use the provided cluster startup scripts as usual. They will start a HA-cluster. Keep in mind that the ZooKeeper quorum has to be running when you call the scripts.
If you don't have a running ZooKeeper installation, you can use the helper scripts, which ship with Flink.
There is a ZooKeeper configuration template in conf/zoo.cfg. You can configure the hosts to run ZooKeeper on with the server.X entries, where X is a unique ID of each server:
The script bin/start-zookeeper-quorum.sh will start a ZooKeeper server on each of the configured hosts. The started processes start ZooKeeper servers via a Flink wrapper, which reads the configuration from conf/zoo.cfg and makes sure to set some required configuration values for convenience. In production setups, it is recommended to manage your own ZooKeeper installation.
Configure ZooKeeper quorum in conf/flink.yaml:
Configure masters in conf/masters:
localhost localhost
Configure ZooKeeper server in conf/zoo.cfg (currently it's only possible to run a single ZooKeeper server per machine):
Start ZooKeeper quorum:
$ bin/start-zookeeper-quorum.sh Starting zookeeper daemon on host localhost.
Start an HA-cluster:
$ bin/start-cluster-streaming.sh Starting HA cluster (streaming mode) with 2 masters and 1 peers in ZooKeeper quorum. Starting jobmanager daemon on host localhost. Starting jobmanager daemon on host localhost. Starting taskmanager daemon on host localhost.
Stop ZooKeeper quorum and cluster:
$ bin/stop-cluster.sh Stopping taskmanager daemon (pid: 7647) on localhost. Stopping jobmanager daemon (pid: 7495) on host localhost. Stopping jobmanager daemon (pid: 7349) on host localhost. $ bin/stop-zookeeper-quorum.sh Stopping zookeeper daemon (pid: 7101) on host localhost.