BadApple report, not a good week.

[Erick Erickson <er...@gmail.com>]

This is not a good week at all: 
Raw fail count by week totals, most recent week first (corresponds to bits):
Week: 0  had  253 failures Most recent 7 days
Week: 1  had  56 failures   7 days before that
Week: 2  had  66 failures
Week: 3  had  83 failures


Going from 56 failures to 253 is A Very Bad Outcome. IDK whether this is an actual horrible regression or we’re reporting on more runs or what.

This makes me fear the changes I made in SOLR-13952 since many of the failures are in the last day. I’ve decided to roll that back anyway, that effort has gone well past the point of diminishing returns so we’ll see if that magically fixes the failure rate.

I’m still going to cull the gradle_8 changes for substantive changes and the suppress warnings zombie threads and push those in the next few days.

Full report attached
Erick