Repository navigation
Conversation
…ithin their own limits A condition, branch, loop or parallel node got the same default 30s limit as a single step, measured over everything inside it. A branch holding an agent step allowed several minutes failed at 30s. Containers now get no default limit: each inner step keeps its own, and an explicit `timeout` on the container still caps the whole group. The per-attempt timer is also cleared once a node settles.
pallaoro
force-pushed
the
container-node-timeout
branch
from
September 26, 2026 19:50
dc64b86 to
1742d12
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
runWithRetryapplies the default per-node limit (maxNodeDurationMs, 30s) to every node without its owntimeout, includingcondition,branch,loopandparallel. For those, the limit is measured over everything inside them. So a condition whose branch holds anagentstep withtimeout: "300s", or a loop of steps that each allow 60s, fails withNode "<container>" timed out after 30000mseven though no inner step exceeded its own limit.Change
timeout(or the default), so nothing runs unbounded.timeouton a container still caps the whole group, as before.SKILL.mdand themaxNodeDurationMsdoc comment say this.Tests
Four new cases in
tests/core.test.ts("container node timeouts"): a condition and a loop whose inner step outlasts the default limit now complete (both failed before the change); an explicit containertimeoutstill fails the run; a single step still gets the default limit. Full suite: 251 pass.Not in this PR
When a node times out, the race rejects but the node's work is not cancelled and keeps running with the run already marked failed. Tracked separately.