Add adaptive provider retry experiment - #500
Draft
DarlingtonDeveloper wants to merge 3 commits into
Draft
DarlingtonDeveloper wants to merge 3 commits into
DarlingtonDeveloper wants to merge 3 commits into
Conversation
DarlingtonDeveloper
added
the
type: improvement
…er-retry # Conflicts: # src/core/agent/runtime/orchestrator.zig
DarlingtonDeveloper
marked this pull request as ready for review
August 31, 2026 16:23
DarlingtonDeveloper
marked this pull request as draft
September 18, 2026 13:51
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add an opt-in
adaptive_v1provider-retry policy forfx ask, with a three-attempt response budget and escalating 30/60/120-second response-head deadlines. Exhaustion persists a paused recovery checkpoint; a successful retry returns normally.Updated against main at
c5cc64cb5518ef8e92f59e3ce2caf32f13d5f860. Main now retries transient failures patiently, so the strict experimental budget is explicitly passed to its current recovery policy. With the experiment off, patient response-head waiting and the ten-minute mid-stream stall window are preserved. Connectivity and liveness probes remain separate from billable retries.Validation: local ReleaseSafe build; focused retry-policy and transport-timing tests; built-binary E2E verifies exactly three failed attempts before pause and successful recovery on the third attempt (16 assertions); formatting and public-surface checks. Exact-current-commit Full CI remains required.
Historical August benchmark snapshots are not results for this updated implementation. A new full-suite hosted matrix is prepared; no current benchmark improvement is claimed.