----- Original Message ----- > From: "Jeff Darcy" <jdarcy@xxxxxxxxxx> > To: "Joseph Fernandes" <josferna@xxxxxxxxxx> > Cc: "Gluster Devel" <gluster-devel@xxxxxxxxxxx>, "Atin Mukherjee" <amukherj@xxxxxxxxxx> > Sent: Monday, February 29, 2016 6:48:35 AM > Subject: Re: tests/basic/tier/tier.t failure in NetBSD > > > Well I will not agree on the part of "The whole ship floats or sinks > > together", > > as in the past we have seen tests(not only with tiering but other features) > > that were failing and were addressed and fixed. > > I'm not sure what you're trying to say here. If a test is failing > because of one line we can change the code, change the failing line, or > even ignore the entire test. That's in approximate order of preference > IMO, but no matter which we do aren't we back to getting the "whole > boat" to float? If we decide that the status-reporting lines in tier.t > aren't important enough to change the result of the test as a whole, > then I think it's perfectly valid to change those lines from "TEST X" to > just "X" or delete them entirely. I'm just saying that we need to > express that decision in code so that our automated test processes will > get what we deem to be the correct answer, instead of letting it get the > "wrong" answer and require a manual override. > I am not saying to manipulate the test so that WRONG results are accepted! Tests are written to detect regressions in any code that is added newly to the system and if the test fails we need to see what is that we have added to the code that is breaking the test. Hope we are on the same page on this. ~Joe _______________________________________________ Gluster-devel mailing list Gluster-devel@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-devel
