Hi folks,

I'd like to bring PHOENIX-2887 to your attention. It means that Phoenix
cannot be used from Yarn applications in a number of scenarios (see history
around the related HBASE-11118), and is bad enough that I think it's
reasonable justification for a patch release. We haven't done a lot of
patch releases, I assume primarily because of a lack of committer bandwidth.

Since I'll need this patch for my own deployment, will effectively be doing
RM duties internally anyway. I'll volunteer to RM a 4.7.1.

Do folks agree that this is bad enough for users to justify a 4.7.1
release? Are there other BUGs we'd like to ship sooner than 4.8.0? How
would an rc0 landing around May 30th sound? Does that compete with some
timeline for 4.8.0? (I've not seen a timeline, but I haven't been watching
closely either)

Thanks,
Nick

Reply via email to