Spark documentation says that for scheduler allocation file we can: put a file named fairscheduler.xml on the classpath
I have included fairscheduler.xml in the root of the jar file that I use with spark-submit but pool scheduler is not used.
Does anyone where to place fairscheduler.xml to be on the classpath? Thx
spark.scheduler.mode=FAIRand works fine locally withfairscheduler.xmlin the resource directory, but when I upload the jar to Databricks it does not work. I am wondering if it's a Databricks bug. Thxspark.scheduler.mode=FAIRfor Databricks as well, but fairscheduler.xml is not getting used from the .jar file and trying to reference it using workspaceconfig("spark.scheduler.allocation.file","file:/Workspace/init/fairscheduler.xml"), it throws a token error. :(