Repository navigation
fix(session): recover interrupted V4 model streams safely - #175
Merged
Merged
Conversation
Recognize gateway transport failures and cap retries before tool execution, preserving earlier parts. The focused SDK regression fails on main and passes with recovery, exhaustion, action and cancellation guards.
sarath-menon
marked this pull request as ready for review
October 7, 2026 13:23
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
All reported issues were addressed across 9 files
Reply with feedback, questions, or to request a fix.
View guided diff | Turn on auto-fix | Re-trigger cubic
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Interrupted V4 model streams were treated as terminal failures even when retrying was safe. This recognizes transport-error envelopes, limits recovery to three total attempts, preserves completed output and stops retries once a tool call is observed or execution begins. The unchanged real-SDK regression went from five failures on baseline to seven passing cases, and package and pre-push typechecks passed (regression evidence). Cloud PR 6394 is merged, and its GitHub staging workflow deployed this release to worker 293. The post-merge live run passed binary verification, screenshots, shell execution, JSON output, cleanup and the complete health gate (run logs). The live run verifies deployed compatibility; upstream connection failures remain unexplained.