[ https://issues.apache.org/jira/browse/KAFKA-4084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16648581#comment-16648581 ]
Jiaxin YE commented on KAFKA-4084: ---------------------------------- Is this issue really resolved? I am using Kafka 1.1.0 and my cluster with 97 hosts and 22,500 partitions face the same problem. Here is what I see in the log for one partition. 1. Broker A went offline for some reasons (deployment, network glitch .. ect) 2. Controller elected broker B as the leader for that partition and let B become the leader. 3. Broker A went online again. 4. Controller let A become the follower of broker B and A is in sync with broker B. 5. Controller triggered auto rebalance and elect A as the new leader and inform B to become the follower. 6. Broker B became the follower and then Controller informed A to be the leader. 7. Broker A became the leader. The problem is the broker A has 100% CPU utilization on step 6 and takes a long to for it to handle the controller request to become a leader. There a great latency between 6 and 7 which cause a partition to live without a leader for around 1- 30 minutes. > automated leader rebalance causes replication downtime for clusters with too > many partitions > -------------------------------------------------------------------------------------------- > > Key: KAFKA-4084 > URL: https://issues.apache.org/jira/browse/KAFKA-4084 > Project: Kafka > Issue Type: Bug > Components: controller > Affects Versions: 0.8.2.2, 0.9.0.0, 0.9.0.1, 0.10.0.0, 0.10.0.1 > Reporter: Tom Crayford > Priority: Major > Labels: reliability > Fix For: 1.1.0 > > > If you enable {{auto.leader.rebalance.enable}} (which is on by default), and > you have a cluster with many partitions, there is a severe amount of > replication downtime following a restart. This causes > `UnderReplicatedPartitions` to fire, and replication is paused. > This is because the current automated leader rebalance mechanism changes > leaders for *all* imbalanced partitions at once, instead of doing it > gradually. This effectively stops all replica fetchers in the cluster > (assuming there are enough imbalanced partitions), and restarts them. This > can take minutes on busy clusters, during which no replication is happening > and user data is at risk. Clients with {{acks=-1}} also see issues at this > time, because replication is effectively stalled. > To quote Todd Palino from the mailing list: > bq. There is an admin CLI command to trigger the preferred replica election > manually. There is also a broker configuration “auto.leader.rebalance.enable” > which you can set to have the broker automatically perform the PLE when > needed. DO NOT USE THIS OPTION. There are serious performance issues when > doing so, especially on larger clusters. It needs some development work that > has not been fully identified yet. > This setting is extremely useful for smaller clusters, but with high > partition counts causes the huge issues stated above. > One potential fix could be adding a new configuration for the number of > partitions to do automated leader rebalancing for at once, and *stop* once > that number of leader rebalances are in flight, until they're done. There may > be better mechanisms, and I'd love to hear if anybody has any ideas. -- This message was sent by Atlassian JIRA (v7.6.3#76005)