Hi wanglei Can you post how you restart the job ? Thanks, Simon
On 06/25/2019 20:11,[hidden email] wrote:
|
If you manually cancel and restart the job, state is only carried forward if you use a savepoint. Can you check if that is what you are doing? On Tue, Jun 25, 2019 at 2:21 PM Simon Su <[hidden email]> wrote:
|
I start and cancel it just in my intellij idea development environment. First click the run button, then click the red stop button, and then click the run button again. Let me google about the savepoint. Thanks, Lei Wang
|
Hi Lei Wang Actually it will not work, job recovery from checkpoint by using jobid to detect the snapshot directory, but when restart it in intellj and didn’t set any configurations, jobid will regenerate and it will regard as a new job, so you get the null state every time, you can follow by https://ci.apache.org/projects/flink/flink-docs-stable/ops/state/checkpoints.html Thanks, Simon On 06/25/2019 21:43,[hidden email] wrote:
|
Free forum by Nabble | Edit this page |