Incorrect scanner default log dirs on Yarn

[zuston]

Search before asking

• I searched in the issues and found nothing similar.

Fluss version

0.9.0 (latest release)

Please describe the bug 🐞

For the logic in getClientScannerIoTmpDir , when the job runs on Hadoop YARN, Flink’s TMP_DIRS may contain multiple directories. In the current codebase, this value is treated as a single path, which can cause data to be written to an incorrect location and lead to uncontrolled disk usage.

Solution

No response

Are you willing to submit a PR?

• I'm willing to submit a PR!