It sounds like most found the plain text email that Josh sent out yesterday
palatable, so I'll do the same. As always, feedback is welcome.
testall:
All pass! (again)
=====================================
dtest:
scrub_test.TestScrubIndexes.test_standalone_scrub
CASSANDRA-12337, new flaky Linux failure on a test with known failures
on Windows. I created a new ticket to track this likely distinct problem.
Stacktrace in the error message makes this look like a C* problem. If you
are familiar with the standalone scrub utilities and want to take a look at
this, assign yourself. Otherwise, I'll find an assignee tomorrow.
cql_tracing_test.TestCqlTracing.tracing_default_impl_test
CASSANDRA-11465, known historically flaky test. Stefania Alborghetti as
assignee, Paulo Motta for review. Patches are being reviewed and tested,
forward progress.
rebuild_test.TestRebuild.simple_rebuild_test
CASSANDRA-11687, Yuki Morishita is assignee. Paulo Motta for review,
dtest runs with the fix look clean. This should be ready to merge soon.
=====================================
novnode_dtest:
rebuild_test.TestRebuild.simple_rebuild_test
Same failure as in the vnode run. CASSANDRA-11687.
=====================================
dtest_upgrade:
Still a lot. (64, up from 53). CASSANDRA-12236 is likely responsible for
most of these, delaying further analysis of upgrade tests until it is
resolved. This ticket is patch available and review in progress by Aleksey
Yeschenko.
=====================================
The dtest_upgrade tests had to be restarted because a machine rebooted.
Otherwise, the test environment looked stable today with no timeouts. The
unit tests in testall continue to pass; forward progress is being made on
the known dtest failures. One new flaky dtest failure today that looks like
it might be a Cassandra issue. Keep up the good work! The signal to noise
ratio is definitely improving.
