Status

Current state: Under Discussion Accepted

Discussion thread: here

JIRA: here

Please keep the discussion on the mailing list rather than commenting on the wiki (wiki discussions get unwieldy fast).

...

The client left as is risks imposing heavy load on ZooKeeper. ZooKeeper itself only has a coarse-grained throttling mechanism in place through its zookeeper.globalOutstandingLimit config which defaults to 1000. This config is insufficient for several reasons:1.

1. the limit is meant to protect ZooKeeper from memory pressure associated with a backlog of requests.

...

1. the limit is applied across all connections. Even with this config, one misbehaved client will affect the other clients.

We need a throttling mechanism in the client-side to give administrators control over Kafka's impact on ZooKeeper, and for this we propose a new broker config called zookeeper.max.in.flight.requests.

...

This config must be set to at least one1.

The default value is set to 10. We ran experiments showing the impact of zookeeper.max.in.flight.requests on completion times for various ZooKeeper-intensive controller protocols. The default was chosen to be the smallest number beyond which the experiment results have found diminishing returns.

...

1. Set the default value to be unbounded (Integer.MAX_VALUE), effectively stating that no such limit be applied and to pipeline as aggressively as possible. This was rejected because a config should not default to removing control over the system.
2. Set the default value to one1, effectively disabling pipelining and maintaining Kafka's synchronous requests to ZooKeeper. This was rejected because we want users to see the benefits of pipelining.