Github user shaneknapp commented on the issue:
https://github.com/apache/spark/pull/21584
just wanted to briefly chime in:
TL;DR: this build will fail until the PRB is running on our ubuntu build
nodes.
we are currently blocked from testing this stuff w/the current iteration of
the PRB on our jenkins for a couple of reasons:
1) the current pull request builder runs only on our old, ancient and
crufty centos 6.9 hosts
2) these hosts have old, ancient and crufty installations of R (3.1.1) and
associated libraries
3) trying to re-create (2) on our shiny, new ubuntu 16.04 boxen has, so
far, been an exercise in futility. :\
4) this is why we currently have a limited number of ubuntu boxes, and also
why things like testing for spark/k8s integration requires an addition build
*on top* of the standard PRB build to make things work.
i'm working now w/some folks at databricks (@weshsiao) to get this under
control enough so that we can get things moving to the ubuntu hosts ASAP.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
