Hi, “uid” is mainly useful when you upgrade your application. It’s used to match the operator state stored in the savepoint. As suggested in [1], “it is highly recommended to assign unique IDs to all operators of an application that might be upgraded in the future.”
[1]. https://ci.apache.org/projects/flink/flink-docs-stable/ops/upgrading.html#matching-operator-state From: "[email protected]" <[email protected]> Date: Thursday, 24 October 2019 at 11:31 PM To: "[email protected]" <[email protected]>, "[email protected]" <[email protected]> Cc: "[email protected]" <[email protected]> Subject: RE: Does operator uid() have to be unique across all jobs? Hi, I have some simple questions on the uid as well. 1. Do we add a uid for every operator e.g. print(), addSink and addSource? 2. For chained operators, do we need to uids for each operator? Or just the last operator? e.g. .map(....).uid("some-id").print().uid("print-id"); Regards, Min From: John Smith [mailto:[email protected]] Sent: Donnerstag, 24. Oktober 2019 16:32 To: Dian Fu Cc: user Subject: [External] Re: Does operator uid() have to be unique across all jobs? Ok cool. Thanks BTW this seems a bit cumbersome... .map(....).uid("some-id").name("some-id"); On Wed, 23 Oct 2019 at 21:13, Dian Fu <[email protected]<mailto:[email protected]>> wrote: Yes, you can use it in another job. The uid needs only to be unique within a job. > 在 2019年10月24日，上午5:42，John Smith > <[email protected]<mailto:[email protected]>> 写道： > > When setting uid() of an operator does it have to be unique across all jobs > or just unique within a job? > > For example can I use env.addSource(myKafkaConsumer).uid("kafka-consumer") in > another job?
