[Issue 213] Fix data race when accessing partition producer state

[cornelk]

Fixes #213

Motivation

Panic due to a race condition.

Modifications

• access to state of partitionProducer is using atomic operations now to be thread-safe
• partitionConsumer will probably need a similar fix as well

Verifying this change

• Make sure that the change passes the CI checks.

• go test -race -count 100 -v -run 'TestDLQ$' . does not panic anymore

This change is already covered by existing tests, such as TestDLQ.

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): (no)
• The public API: (no)
• The schema: (no)
• The default values of configurations: (no)
• The wire protocol: (no)

Documentation

• Does this pull request introduce a new feature? (no)

[jiazhai]

Thanks for the great help @cornelk.
@cornelk @wolfstudy Do you think it is possible to add some test like:
"go test -race -count 100 -v -run 'TestDLQ$' ",
into the github action CI process?

[wolfstudy]

Thanks for the great help @cornelk.
@cornelk @wolfstudy Do you think it is possible to add some test like:
"go test -race -count 100 -v -run 'TestDLQ$' ",
into the github action CI process?

Good idea, will add this to ci

[wolfstudy]

LGTM +1

[wolfstudy]

ping @merlimat

@cornelk can you help to check whether this problem also exists on the consumer side?