Flink CDC Issue Import created FLINK-34796:
----------------------------------------------
Summary: [Bug] flink cdc has lot of lamda class, it may leaked in
flink standalone cluster after task start and stop many times
Key: FLINK-34796
URL: https://issues.apache.org/jira/browse/FLINK-34796
Project: Flink
Issue Type: Bug
Components: Flink CDC
Reporter: Flink CDC Issue Import
### Search before asking
- [X] I searched in the
[issues|https://github.com/ververica/flink-cdc-connectors/issues] and found
nothing similar.
### Flink version
1.16.0
### Flink CDC version
2.2.1
### Database and its version
8.0
### Minimal reproduce step
-deploy a flink standalone cluster
- start a flink cdc task and stop the task three or more times
### What did you expect to see?
- the cluster could unload the related class in the task and the metaspace
could decrease after full gc
### What did you see instead?
- the metaspace of flink cluster always increase until oom
### Anything else?
i dump the heap and analyse with jprofile, the ChildFirstClassLoader could not
be unload because many lambda class
![Uploading image.png…]()
could someone help me give some advice, the lambda in java is common, and flink
cdc used lot of lambda, i tried with a small demo, only lamda, but it could be
unload when full gc.
### Are you willing to submit a PR?
- [x] I'm willing to submit a PR!
---------------- Imported from GitHub ----------------
Url: https://github.com/apache/flink-cdc/issues/2196
Created by: [Fernandoize|https://github.com/Fernandoize]
Labels: bug,
Created at: Sat Jun 10 22:41:42 CST 2023
State: open
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
