You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[P2] Handle tuple request timeouts consistently in pod exec #18
Passing _request_timeout=(connect_timeout, read_timeout) to pod exec is reported to fail with a float-versus-tuple comparison error. The REST transport supports a timeout tuple, but the WebSocket path passes the value into its run loop without equivalent handling.
Reproduce on the current client and define supported connect/read timeout semantics for exec. Do not silently drop one tuple component or conflate connection timeout with command duration.
Acceptance criteria:
Add focused tuple-timeout regression coverage.
Preserve supported scalar/default timeout and non-preloaded streaming behavior.
Document any transport-specific constraints and surface unsupported values explicitly.
Clarification to the original assignment: this task's designated agent/base branch is agent/p2-issue-18. Raise/target the PR against that branch, not master; the existing draft PR has been retargeted accordingly. Keep the requested commit order: first a test-only commit demonstrating the issue, then a separate fix commit only if reproduced. If it does not reproduce, leave the evidence/test commit and explain why no fix is appropriate.
Activity
brendandburns commentedon Nov 19, 2026
Clarification to the original assignment: this task's designated agent/base branch is
agent/p2-issue-18. Raise/target the PR against that branch, notmaster; the existing draft PR has been retargeted accordingly. Keep the requested commit order: first a test-only commit demonstrating the issue, then a separate fix commit only if reproduced. If it does not reproduce, leave the evidence/test commit and explain why no fix is appropriate.