Tim Quinn

Resolved with a separate push.

Yes, exactly. That avoids lots of duplicate-metric-registered errors. The `@HelidonTest` annotation accepts `resetPerTest` which restarts the Helidon container for each test, but with that specified the test container is documented...

@Captain1653 The ports to 2.x and 4.x will need to be in separate PRs; a single PR can merge only into one branch. I've added check boxes for each Helidon...

Do we need both "in" and "not in" for this? I would think that either * one can be derived or inferred from the other, or * the names do...

I see. I know we prefer short names, but if that _means_ "forbidden" maybe we should call it that or "excluded" or something else instead of "not in" to avoid...

At the moment, we do not have an ETA for this fix. I wish we had more definite news to share on this.

+1 with an Intel Mac with Ventura 13.2.1 and Java 20 Update. Every run was failing at this test, even when I dropped back to Java 19. I then happened...

I updated the samples locally with no change to the spring ones but I rebased and pushed anyway. This time there is one failed spring job instead of two, again...

With Helidon SE there is “no magic” and the developer has the flexibility (and therefore the responsibility) of complete control. You need to prepare some context so that the Helidon...

@tomas-langer and @Verdent Is there any reason we _should not_ use the currently-active span as the parent in this use case, assuming that the span context is absent from the...