Details
-
Improvement
-
Status: Resolved
-
Minor
-
Resolution: Incomplete
-
None
-
None
Description
Many custom Hadoop Input/Output Formats register metrics via Hadoop's Counter infrastructure. For example, the Input/Output formats offered by es-hadoop registers quite a few handy metrics.
It'd be very helpful for something like the SparkHadoopWriter class (not sure of the equivalent on the read side) to automatically discover and register all Hadoop counters as Spark metric sources.