Re: Minimizing global store restoration time

Hi Taylor, 1) Yes we do allow users to have separate config values for global consumers / restore consumers via StreamsConfig#restoreConsumerPrefix and StreamsConfig#globalConsumerPrefix, as Patrik pointed out. 2) I think I agree with you that for global consumer, it is worth while to allow one

Re: TimestampExtractor based on Postgres Transaction IDs for doing SessionWindow joins

Hello Ashhar, Could you elaborate a bit more on why you'd like to perform a stream-stream join than a stream-table join, or even a table-table join with Streams? What's the joining streams for, are they both changelog streams from Debezium/Postgres? What's the session window needed for? Guozhang

Re: kafka stream. streams.allMetadata() get nothing back.

Hi Nan, You'll need to set the "application.server" property for Kafka Streams. Since you are using the "streams#allMetadata()" method I'm assuming you are trying to set up interactive queries, so here are two links that should help you get started:

Consumer group indefinitely loops in rebalance problem

Hi, I’m using Kafka kafka_2.11-1.0.0 and the java client version is 0.10.1.0. The topic has 9 partitions and there are 2 processes each has 5 consumer threads consuming this topic. The consumer group indefinitely rebalances after the 2 processes started. This is the consumer config info

AW: Configuration of log compaction

Hi kafka users, I still have the problem of the log cleaner dying regularly. Yesterday I restarted 3 of my brokers in a 5-broker cluster. On two of them the log cleaner threads already died again before they were able to clean up most of the old data. A general question in this regard: 1.)

Re: [VOTE] 2.2.0 RC0 [CANCELED]

We discovered a blocker issue: https://issues.apache.org/jira/browse/KAFKA-8012 Thus, I am cancelling the current vote. I will cut a new RC after the blocker is fixed. Nevertheless, please keep testing. If we find something else before the new RC is cut, it would streamline the overall process.