slow consumer recovery after gc pause

Hello, We are running kafka 0.9.0.1 and using the simple consumer to consume from a topic with 8 partitions. The consumer JVM infrequently runs into large gc pauses (60s to 90s, stop the world gc). These gcs are unrelated to kafka. We are usually consuming at 5k messages/sec on the topic. Rig

Re: Partial broker shutdown causing producers to stall even with replicas

even if one ZK node is bad is especially terrible. On Tuesday, November 17, 2015, Mahdi Ben Hamida wrote: Hello, See below for my original email. I was wondering if anybody has feedback on the 4 questions I've asked. Should I go ahead and file this as a bug ? Thanks. -- Mahdi. On 11/12/

Re: Partial broker shutdown causing producers to stall even with replicas

Hello, See below for my original email. I was wondering if anybody has feedback on the 4 questions I've asked. Should I go ahead and file this as a bug ? Thanks. -- Mahdi. On 11/12/15 2:37 PM, Mahdi Ben Hamida wrote: Hi Everyone, We are using kafka 0.8.2.1 and we noticed that

Partial broker shutdown causing producers to stall even with replicas

Hi Everyone, We are using kafka 0.8.2.1 and we noticed that kafka/zookeeper-client were not able to gracefully handle a non existing zookeeper instance. This caused one of our brokers to get stuck during a shutdown and that seemed to impact the partitions for which the broker was a leader even