paul-rogers commented on PR #12368:
URL: https://github.com/apache/druid/pull/12368#issuecomment-1120040331

   @kfaraz, thank you for your thorough review, and for trying out the new 
setup. Always great to know it runs on a machine other than my own!
   
   You mentioned flaky test and how to retry them. Two thoughts on that.
   
   First, we should not have flaky tests. IMHO, such tests either:
   
   * Are flaky because they start running before the cluster is stable,
   * Are not telling us anything if the test themselves are flaky (because they 
depend on timing, or on behavior which is inherently non-deterministic, such as 
the ordering of events from different services.)
   * Are point out actual issues with Druid: that clients would have to retry 
operations. We should either a) fix that issue, or b) document it. Either way, 
the tests should be prepared for whatever race or non-deterministic condition 
is in question.
   
   The new framework eliminates the first issue. The framework ensures that 
services are ready before launching tests. This means that either the test or 
Druid is flaky. Either way, we should fix he issue: remove the test if it is 
not useful, else fix it or fix Druid (perhaps adding a way to synchronize when 
needed for testing.)


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to