Re: Streams changelog topic retention is high

2017-10-31 Thread Sameer Kumar
Thanks Damian, I worked. I changed the StreamsConfig.WINDOW_STORE_ CHANGE_LOG_ADDITIONAL_RETENTION_MS_CONFIG and was able to reduce the retention time of the changelog topic. -Sameer. On Mon, Oct 30, 2017 at 9:38 PM, Damian Guy wrote: > The retention for the joins is as specified above. With

after add broker how to change partition leader to new broker

2017-10-31 Thread lk_kafka
hi,all: on my cluster , some topic Under-replicated 100% , I have 7 broker and the topic have 7 partition ,but the topic now only use 5 broker as leader , how can I chang it to use all the broker as leader ? 2017-11-01 lk_kafka

kafka user

2017-10-31 Thread Karthigeyan
Hi ,     pls add to the user group. Thanks , Karthigeyan

KAFKA For System z

2017-10-31 Thread Sultan S. Alsaleem
Dear , I just want to know if KAFKA is supported on IBM System z Linux or z/OS . Regards,

Kafka performance

2017-10-31 Thread Oussema Ben lamine
Hello, I am a new kafka user, trying to reach kafka limits :1 000 000 messages per second. Therefore, I loaded three Azure VMs having this configuration: 16 CPU Cores 56 Go Ram All disks are premium LRS and minimum capacity 512 Go. Download speed 2696.28 Mbit/s Upload speed 1121.13 Mbi

ZkUtils.getAllPartitions giving more partition counts.

2017-10-31 Thread satyajit vegesna
Hi, Would like to understand the purpose of ZkUtils.getAllPartitions, as when i try to use the method, i end up getting wrong number of partitions assigned to Topics, i am not really sure if my understanding is wrong about this method. i have assumed this method would return the partition count. B

Re: Re: Servers Getting disconnected

2017-10-31 Thread Ted Yu
How about upgrading to 0.10.1.1 or higher, as suggested by Ismael ? On Tue, Oct 31, 2017 at 3:42 AM, Yuanjia wrote: > Hi Ted, > It doesn't look like the same issue. > In my case, the node 6 doesn't shrink for the partitions it owns the > ISR's down to itself, and all clients work well. >

Re: regarding number of Stream Tasks

2017-10-31 Thread pravin kumar
ohhh...thank you. Its cleared now On Tue, Oct 31, 2017 at 4:36 PM, Damian Guy wrote: > Hi, the `map` when it is followed by `groupByKey` will cause a > repartitioning of the data, so you will have your 5 tasks processing the > input partitions and 5 tasks processing the partitions from the > rep

Re: regarding number of Stream Tasks

2017-10-31 Thread Damian Guy
Hi, the `map` when it is followed by `groupByKey` will cause a repartitioning of the data, so you will have your 5 tasks processing the input partitions and 5 tasks processing the partitions from the repartitioning. On Tue, 31 Oct 2017 at 10:56 pravin kumar wrote: > I have created a stream with

Re: regarding number of Stream Tasks

2017-10-31 Thread pravin kumar
I have created a stream with topic contains 5 partitions and expected to create 5 stream tasks ,i got 10 tasks as 0_0 0_1 0_2 0_3 0_4 1_0 1_1 1_2 1_3 1_4 im doing wordcount in this example, here is my topology in this link: 1. https://gist.github.com/Pk007790/72b0718f26e6963246e83da992

Re: Re: Servers Getting disconnected

2017-10-31 Thread Yuanjia
Hi Ted, It doesn't look like the same issue. In my case, the node 6 doesn't shrink for the partitions it owns the ISR's down to itself, and all clients work well. I notice KAFKA-5153, it maybe same to mine. But it doesn't update for a long time. From: Ted Yu Date: 2017-10-30 17