LogTieringTask#commitRemoteLogManifest() may block forever when coordinator failover

[luoyuxia]

Search before asking

• I searched in the issues and found nothing similar.

Fluss version

0.6.0 (latest release)

Please describe the bug 🐞

LogTieringTask#commitRemoteLogManifest() may block forever when coordinator failover.
It'll call method coordinatorGateway.commitRemoteLogManifest(request).get(), when coordinator server failover, the get() may block forever and will never return.

Solution

TBH, I'm still wondering wy it'll block forever. Theoretical, when the coordinator server failover, the connection will break and cause the client return with a disconnection exception. But it may happen that it'll block and never return from phenomenon.

Anyway, #279 will introduce global timeout to solve the problem.

Are you willing to submit a PR?

• I'm willing to submit a PR!