Skip to content

Resume on a full fleet fails with ResourceExhausted, not Aborted - #2846

Open
Eric Curtin (ericcurtin) wants to merge 3 commits into
agent-substrate:mainfrom
ericcurtin:exhausted-assignment-outcome
Open

Eric Curtin (ericcurtin) wants to merge 3 commits into
agent-substrate:mainfrom
ericcurtin:exhausted-assignment-outcome

Conversation

@ericcurtin

Copy link
Copy Markdown
Contributor

Fixes #2259

The last assignment attempt decides: a full worker is ResourceExhausted and no_capacity, a version conflict stays Aborted.

  • Tests pass
  • Appropriate changes to documentation are included in the PR

…ies run out

The last attempt now decides: a full worker is no_capacity, a version conflict stays Aborted.

Signed-off-by: Eric Curtin <eric.curtin@docker.com>
@ericcurtin

Copy link
Copy Markdown
Contributor Author

Rebased onto main, e51492d. Luiz Oliveira (@laoj2) Dmitry Berkovich (@dberkov) PTAL when you get a chance. Thank you!

Signed-off-by: Eric Curtin <eric.curtin@docker.com>
…orker

BindActorToWorker reports a missing Actor as ErrNotFound now.

Signed-off-by: Eric Curtin <eric.curtin@docker.com>

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

area/api-machinery area/scheduling kind/bug Something isn't working / bugfixes

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Bug]: A full fleet reports Aborted instead of ResourceExhausted when the assignment retries run out

2 participants