Repository navigation
spec(queen): dispatch_exit -- a provider refusal never charges the issue (Closes #10008) - #10011
Merged
Merged
Conversation
…sue (Closes #10008) The bee actors in gHashTag/trios ask counts_against_issue whether an end spends a counted run. A provider error did, so three quota refusals in a row gave the issue up, while the next dispatch would take another key. key_failover.t27 charges_issue (#9993) charges only a bad request; the two cards now agree. A bad request (HC_4XX) still gives up at once, and a refusal that keeps repeating still ends at STREAK_LIMIT. Railway lab: test-report 11/11, 0 vacuous, 4 invariants; tri mutate spec 81/84 killed, 3 equivalent survivors in the untouched restart_wait_seconds clamps. The wasm recipe rebuilds the pinned 31c0e524 from the old card; the new card gives b990a601 on two builds. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
gHashTag
enabled auto-merge (squash)
October 17, 2026 16:13
Contributor
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #10008. Part of #9976. Related: #9993, #7851.
Why
The bee dispatcher as keyed actors (gHashTag/trios
queen-bee-actors.ts) askscounts_against_issuewhether an end spends a counted run of the issue.RUN_LIMIT) gave the issue up. The next dispatch would have taken another key.key_failover.t27charges_issue(queen: bee key failover -- on 429/1305 the next attempt goes to another key, then the other pool, instead of retrying the same key until the bee dies #9993) charges only a bad request. This card disagreed with it, and this PR makes the two agree.This is the step that comes before switching trios to
TRIOS_QUEEN_DISPATCH=actors.The change
counts_against_issue(DX_PROVIDER_ERROR)is now false.HC_4XX) still gives up at once.STREAK_LIMIT(6).a_provider_refusal_never_charges_the_issue. It also pins that a timeout and an unknown end remain counted runs.Evidence (Railway lab, master de2cff8)
t27c test-report: 11 tests, 0 FAIL, 0 vacuous, 4 invariants. The seal is regenerated.tri mutate spec: 81 of 84 mutants killed. The 3 survivors are equivalent mutants in the untouchedrestart_wait_secondsclamps (lines 192, 198 and 202). At the boundary both branches return the same value.specs/PINrecipe:31c0e524....b990a601...on two builds.Next
The trios vendoring PR re-vendors the
.t27, the.wasmand thespecs/PINlines. It changes no TypeScript.🤖 Generated with Claude Code