問題文
An engineer needs to read the standard error output of a specific executor to investigate an out-of-memory failure. Where is that reachable from in Apache Spark?
選択肢
- From the executor listing in the web interface, which carries links to the standard error and standard output of each executor.
- From the SQL listing, which shows the per-query logs of each executor.
- From a fixed temporary directory on the machine that hosts the driver, and every executor writes into it over the shared file system.
- From the submit command with a verbose flag, which prints the executor logs to the console of the submitting shell so that they can be captured together with the driver output in a single stream.