Reputation: 101
I have an EMR cluster in AWS.
I am building a RStudio Server to use that cluster.
I am able to get a spark-shell and execute spark.sql("show tables").show
and see the tables in HDFS from the server that is running RStudio.
However if I execute spark.sql("select * from ny_taxi_hdfs limit 10").show
it fails with the error.
Initial job has not accepted any resources.
If I do the same thing from the master it works fine.
It does look like it is using yarn "(master = yarn, app id = application_1568817071939_0034)"
Not even sure where to look.
Upvotes: 1
Views: 155
Reputation: 101
Firewall/Security Groups. Once I opened it up everything worked. I hate when I ask a question and a day later answer it.
Upvotes: 1