[ https://issues.apache.org/jira/browse/CASSANDRA-12277?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15391446#comment-15391446 ]
Sylvain Lebresne commented on CASSANDRA-12277: ---------------------------------------------- I'll admit that at face value that seems a weird and potentially dangerous thing to do to me. I can easily see us defaulting to such mechanism every time we get a flaky test and that hiding genuine problems more often than not. I don't know the details about that specific {{ReplicationAwareTokenAllocatorTest.testNewCluster}} test, but if a test knowingly fails a number of the times, this feels to me like the test isn't precise enough and should be improved. Note that I can buy that exceptionally a test is such that making it pass 100% of the time is too much work to be worth the trouble but the test is still nice to have, but that should be exceptional and I'd rather handle that in a case-by-case basis, with very precise comment as to why this is allowed to fail some of the times, to make it clear it's a special case. > Extend testing infrastructure to handle expected intermittent flaky tests - > see ReplicationAwareTokenAllocatorTest.testNewCluster > --------------------------------------------------------------------------------------------------------------------------------- > > Key: CASSANDRA-12277 > URL: https://issues.apache.org/jira/browse/CASSANDRA-12277 > Project: Cassandra > Issue Type: Bug > Reporter: Joshua McKenzie > Assignee: Branimir Lambov > Priority: Minor > Labels: test > > From an offline discussion: > bq. The ReplicationAwareTokenAllocatorTest.testNewCluster failure is a flake > -- randomness will sometimes (on the order of 1/100) cause it to fail. > Extending the ranges to avoid these flakes goes too far and makes the test > meaningless. > bq. How about instead of @flaky/@Ignore which currently indicates a test that > intermittently fails but we do not expect it to, we instead use @tries, or > @runs, or some annotation that indicates "run this thing N times, if M pass > we're good". This would allow us to keep the current "we don't care about > these test results (in context of green test board) because intermittent > failures are not expected and the test quality needs shoring up" from "we > expect this test to fail sometimes in this particular way." -- This message was sent by Atlassian JIRA (v6.3.4#6332)