I think I can set . a new uuid but it seems `allowNonRestoreState` is a CLI
hint. I need the "automatic" restart on failure to use the new uuid.
Our use case has no use of data on Kafka that is not current.
On Thu, Nov 22, 2018 at 11:16 PM Tony Wei wrote:
> Hi Vishal,
>
> AFAIK, the current beha
I'm using Flink 1.4.2 and deploying job on Yarn Cluster.
I have a streaming job, which flattens the data and outputs it. It basically
takes a input record and produces n output record. I'm using Table Function
for this. The logic to flatten the data is implemented in a UDF. The UDF has
a counter
Hi,
I want to do window join on multiple Kafka streams (say a, b, c) on common
field in all 3 streams and apply some custom function on joined stream. As
I understand we can join only 2 streams at a time via DataStream api. So
may be I need to join a and b first and then join first joined stream wi
