Re: kafka-streams batch restore state

[Matthias J. Sax]

`kill -9` will force-stop the application the hard way, thus, it cannot
write a checkpoint file that it needs for a clean restart.

Just you should use `kill` (SIGTERM, not SIGKILL == -9) to tell Kafka
Streams to shutdown gracefully. This will allow Kafka Streams to write a
checkpoint file. This way, on restart Kafka Streams will pick up the
state correctly.


-Matthias


On 11/29/18 6:49 PM, meigong.wang wrote:
> https://cwiki.apache.org/confluence/display/KAFKA/KIP-116%3A+Add+State+Store+Checkpoint+Interval+Configuration
> This kafka KPI helped me. I use ‘kill -9' to stop my application.
> 
> 
> Thanks
> 
> 
> 原始邮件
> 发件人:Matthias J. saxmatth...@confluent.io
> 收件人:usersus...@kafka.apache.org
> 发送时间:2018年11月30日(周五) 10:30
> 主题:Re: kafka-streams batch restore state
> 
> 
> Not sure. Maybe you can provide a larger part of the log? Maybe in DEBUG 
> level? -Matthias On 11/28/18 10:08 PM, meigong.wang wrote:  When I restart my 
> kafka-streams application, sometimes I can see following log：    [Consumer 
> clientId=market-kline-stream-wmg-ticker-d891eeaf-2932-45b8-b7e2-9720904a69a5-StreamThread-3-restore-consumer,
>  groupId=] Resetting offset for partition 
> market-kline-stream-wmg-ticker-KSTREAM-REDUCE-STATE-STORE-0000000003-changelog-0
>  to offset 0.    I am sure the rocksdb directoryexists，I didn't change any 
> configuration，what happened?Is this a bug in kafka-streams，the version is 
> 1.1.1.
> 

signature.asc
Description: OpenPGP digital signature