Flink MapState manages the TTL of each map entry independently. If you
add/update a “new hash”, the “old hash” that was written one hour ago stays
unaffected and can still expire as it should be. So it should meet your
requirements of deduplication and size-control.
> 2025年6月4日 10:54，Sachin Mi
Hi Nate, Hi Lasse,
FYI, if you're still interested in the test harness for KeyFunctions with State
V2, please note that the ticket [1] mentioned earlier has been resolved.
Although it hasn't been officially released yet, you can try out the new
harness methods. These methods can be found in the
Hi patricia,
```setPredefinedOptions ``` can still be found in EmbeddedRocksDBStateBackend
in Flink 2.0.
You can set the option programmatically via config:
config.set(
RocksDBOptions.PREDEFINED_OPTIONS,
PredefinedOptions.FLASH_SSD_OPTIMIZED.name());
Best,
Han Yin
> 2025年7月15日 13
ink/flink-docs-release-2.0/docs/dev/datastream/fault-tolerance/state_v2/>
Sincerely,
Han Yin
> 2025年6月30日 14:49，Shikhar Raje 写道：
>
> Hi Yin,
>
> I understand that #2 (adding `enableAsyncState()`) was the fix that I was
> looking for. We're in the process of bu
Hi, Shikhar
It seems you have used state V2 APIs in your code. Please check the imports in
your ```FraudDetector.java``` and you may see that ```ValueState``` and
```ValueStateDescriptor``` are under ```org.apache.flink.api.common.state.v2```
package.
About your questions:
1. The asynchronous be
