September 2018 at 11:15
To: Oleksandr Nitavskyi
Cc: R&D/Product Engineering/PRIME/Delight ,
"gor...@data-artisans.com" , Juan Gentile
, "user@flink.apache.org"
Subject: Re: Weird behaviour after change sources in a job.
Hi Oleksandr,
The mapping of state to operator is d
ndr
>
>
>
> *From: *Dawid Wysakowicz
> *Date: *Thursday, 13 September 2018 at 09:59
> *To: *Juan Gentile ,
> "user@flink.apache.org"
>
> *Cc: *R&D/Product Engineering/PRIME/Delight
> ,
>
> *Subject: *Re: Weird behaviour after change source
gt; *From: *Dawid Wysakowicz
> *Date: *Thursday, 13 September 2018 at 09:59
> *To: *Juan Gentile , "user@flink.apache.org"
>
> *Cc: *R&D/Product Engineering/PRIME/Delight ,
>
> *Subject: *Re: Weird behaviour after change sources in a job.
>
>
>
> Hi
topics which are different from
the currently consumed one.
Thank you
Oleksandr
From: Dawid Wysakowicz
Date: Thursday, 13 September 2018 at 09:59
To: Juan Gentile , "user@flink.apache.org"
Cc: R&D/Product Engineering/PRIME/Delight ,
Subject: Re: Weird behaviour after change so
Hi Juan,
I think this is somehow expected behaviour. Flink, in order to provide
proper processing semantics keeps track of partitions offsets
internally, and checkpoints those offsets. FlinkKafkaConsumer supports
also new partitions discovery. Having in mind both of those features, if
you restart
