Github user mccheah commented on the issue:
https://github.com/apache/spark/pull/21748
> there was a change in Spark recently in how the driver self-discovered
its hostname by default, if I am not mistaken. Can't recall the exact patch. I
remember that change specifically prompting us to set up the headless service
for cluster mode.
Not exactly recently, but at some point after we started work in mainline.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
