On 22 Oct 2015, at 21:54, Chester Chen
<ches...@alpinenow.com<mailto:ches...@alpinenow.com>> wrote:
Thanks Steve
Likes the slides on kerberos, I have enough scars from Kerberos while
trying to integrated it with (Pig, MapRed, Hive JDBC, and HCatalog and Spark)
etc. I am still having trouble making Impersonating to work for HCatalog. I
might send you an offline email to ask some pointers
pointer? Run
Thanks for the ticket.
Looking at the code, it seems to me you could bypass the hive loading by
setting spark.yarn.security.tokens.hive.enabled=false
I've started on a patch, but so far just cleaned up the code ready for writing
a test against it. Expect more tomorrow
https://github.com/apache/spark/pull/9232
