There are many use cases of cron jobs running in distributed environment. Since redundancy is required on production, certain co-ordination is needed to guarantee multiple instances don't fire at once.
ZooKeeper is a distributed, open-source coordination service for distributed applications
As Zookeeper data model is styled after the familiar directory tree structure of file systems they are most suited for this.
I provide here a custom lock implementation (i.e. ZookeeperDistributedLock), using zookeeper proposed guidelines
Acquire Lock
1. Call create( ) with a pathname of "_locknode_/lock-" and the sequence and ephemeral flags set.
2. Call getChildren( ) on the lock node without setting the watch flag (this is important to avoid the herd effect).
3. If the pathname created in step 1 has the lowest sequence number suffix, the client has the lock and the client exits the protocol.
4. The client calls exists( ) with the watch flag set on the path in the lock directory with the next lowest sequence number.
5. if exists( ) returns false, go to step 2. Otherwise, wait for a notification for the pathname from the previous step before going to step 2.
Release Lock
1. The unlock protocol is very simple: clients wishing to release a lock simply delete the node they created in step 1.
This project uses Quartz (scheduler framework) to fire 2 jobs at same time trying to acquire the lock. Since both Jobs create a new instance of the ZookeeperDistributedLock, we can consider this setup to be emulating 2 distributed cron instances.
Since I already use Kafka (which uses Zookeeper for discovery) , I just spawn a container to test the same .
docker pull spotify/kafka
docker run -p 2181:2181 -p 9092:9092 --env ADVERTISED_HOST=localhost --env ADVERTISED_PORT=9092 --name kafka_local spotify/kafka
You can also pull zookeeper image to test the same.
docker pull zookeeper
Run the below commands for starting the app.
Build the executable
gradle fatJar
Run
java -jar build/libs/ZookeeperDistributedLock-1.0.jar