Github user StephanEwen commented on the issue:
https://github.com/apache/flink/pull/2065
You already outlined one of the drawbacks (holding on to the classes).
There are probably other parts in the execution graph that currently hold
onto user defined classes as well.
A good (but more involved) longer term solution would be to create a
"archived execution graph" that can be serialized and that does not contain any
user classes, but only "stringified" versions of configs, accumulators,
metrics, etc.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
