Thank you. For others, the fix assumes that the issue was that the events list had stale events left in it. The list is reset when setLatch is called at the beginning of the test, ensuring all stale data is removed.
Mark On Mon, 17 Jun 2024 at 11:46, Francisco Javier Tirado Sarti < [email protected]> wrote: > Pere has already reviewed it, we are waiting for the result on CI. > > On Mon, Jun 17, 2024 at 12:41 PM Francisco Javier Tirado Sarti < > [email protected]> wrote: > > > I opened PR for this one > > https://github.com/apache/incubator-kie-kogito-apps/pull/2071, > > > > On Mon, Jun 17, 2024 at 12:34 PM Mark Proctor <[email protected]> > > wrote: > > > >> I'm looking at some of the failing tests on Jenkins, and started with > >> EmbeddedJobsServiceTest.testJobService > >> https://github.com/apache/incubator-kie-kogito-apps/pull/2068 > >> > >> This test looks potentially non-deterministic. It has a wait latch that > >> applies the test after 8 events are received. It then gets the list of > >> events, but it's not a safe copy. While it's iterating the stream, it > >> receives another event. It's not clear what this event is or if this > >> additional event is relevant or even potentially a bug. > >> > >> 1) @Ignore the test. > >> 2) Make a thread-safe clone of the list so it's safe to iterate. > >> 3) Update the test to correctly process all events that are received and > >> ensure all relevant logic is tested, and there is no risk of additional > >> events when iterating the stream. > >> > >> Looking at the test, it was added earlier this year by Pere; I might > >> recommend the original author would be best placed to address via > points 1 > >> to 3. > >> > >> Mark > >> > > >
