Github user vanzin commented on the issue:
https://github.com/apache/spark/pull/22614
> a better and safer solution by introducing exponential backoff with
retries
I'm confused about that suggestion. What is being retried? The user query?
The HMS call? What is changing in between the retries that gives any hope that
a future call will succeed? And if all the retries fail, do you fallback like
this change is doing, or what?
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
