[KAFKA-3042] updateIsr should stop after failed several times due to zkVersion issue

• Type: Bug
• Status: Reopened
• Priority: Critical
• Resolution: Unresolved
• Affects Version/s: 0.10.0.0, 2.4.1
• Component/s: controller
• Labels: reliability
• Environment: jdk 1.7; centos 6.4
• Assignee: Dong Lin
• Reporter: Jiahongchao
• Created: 24/Dec/15 10:24
• Updated: 03/Mar/21 15:25

Description

sometimes one broker may repeatly log
"Cached zkVersion 54 not equal to that in zookeeper, skip updating ISR"
I think this is because the broker consider itself as the leader in fact it's a follower.
So after several failed tries, it need to find out who is the leader

Issue Links

• is duplicated by: KAFKA-3577 Partial cluster breakdown
• is related to: KAFKA-12405 Inconsistent broker without active topics still involved in coordinating consumer group
• is related to: KAFKA-1407 Broker can not return to ISR because of BadVersionException
• is related to: KAFKA-2729 Cached zkVersion not equal to that in zookeeper, broker not recovering.
• is related to: KAFKA-1382 Update zkVersion on partition state update failures
• relates to: KAFKA-3577 Partial cluster breakdown