Timur Zununbekov

I think that the idea of per-object delivery was to make it a default config that's applied on every step. And if step has its own delivery, it overrides default...

Doesn't look like I'll have time for this one anytime soon /unassign

@nkreiger Thanks for the PR. Before I review the code, does this change somehow correlate with this https://github.com/triggermesh/triggermesh/pull/1049 ?

@ajp-lsq looks like this condition is left there because triggermesh controller still uses these credentials to access S3 API. As [described](https://github.com/triggermesh/triggermesh/blob/2f78b09ca32bc29c1c6a3cdd4886a8a899216bb0/config/300-awss3source.yaml#L171-L175) in the CRD, the controller also can work with...

> Bring any missing feature to Function. Just remind me what we needed to bring in and I'll do it

I vote for the following: 1. Re-organize hierarchy in [docs](https://github.com/triggermesh/docs) to make it a bit more formal, 2. Move [docs](https://github.com/triggermesh/triggermesh/tree/main/docs) from this repo to the `triggermesh/docs`, 3. This repo can...

For reviewers - see linked PR on how this change is supposed to be used. If there is another option to achieve the same result with less effect on the...

@sixcorners thanks for filing the issue @sameersbn what do you think? I agree that having a root domain that does not resolve to anything is not really useful

Thanks, @sstubbs. We don't have any particular plans for adding new runtimes in the near future as we're focusing more on improving [Functions](https://docs.triggermesh.io/concepts/functions/) based on current KLR runtimes.

@nkreiger thanks for the contribution 🙇‍♂️ By any chance, is it possible to add the test that would reproduce described race condition to make sure that we fixed it? @sameersbn...