Fix per-request SSE reconnection budget on clean EOF - #3409
Conversation
A per-request reconnect that opens successfully, emits an id-bearing priming event, and then reaches EOF without a JSON-RPC response used to recurse with attempt=0, resetting the reconnection budget. A no-timeout request (subscriptions/listen) could therefore reconnect forever instead of resolving its waiter with CONNECTION_CLOSED after MAX_RECONNECTION_ATTEMPTS. The clean EOF path now consumes the same per-request budget as the failed-reconnect path. A regression test drives StreamableHTTPTransport._handle_reconnection() with mock streams that always open fine, emit one id-bearing priming event with empty data, then EOF; the waiter must be resolved with CONNECTION_CLOSED after exactly MAX_RECONNECTION_ATTEMPTS reconnects, and no further request may be sent after the budget is exhausted. Fixes modelcontextprotocol#3307
|
This PR has been closed automatically. This repo only keeps pull requests open when they come from a maintainer, or from a contributor a maintainer has assigned to the linked issue, and you aren't currently assigned to #3307. If a maintainer assigns you to #3307, this PR reopens on its own and there's nothing more you need to do here. Assignment is a maintainer call based on capacity; comments that only ask to be assigned don't factor in. What does help is engaging on the issue itself by confirming the repro, explaining why it matters for your use case, or describing the approach you'd take. You're welcome to keep pushing commits here (just avoid force-pushing, since GitHub can't reopen a rewritten branch), but that on its own won't get the PR reviewed or the issue assigned, and realistically most auto-closed PRs stay closed. There's no need to open a new PR either way. CONTRIBUTING.md has the full reasoning, but in short:
Maintainers: reopen, remove |
Summary
Fix per-request SSE reconnection budget on clean EOF
A per-request reconnect that opens successfully, emits an id-bearing priming event, and then reaches EOF without a JSON-RPC response used to recurse with
attempt=0, resetting the reconnection budget. A no-timeout request (subscriptions/listen) could therefore reconnect forever instead of resolving its waiter withCONNECTION_CLOSEDafterMAX_RECONNECTION_ATTEMPTS.The clean EOF path now consumes the same per-request budget as the failed-reconnect path.
Fixes #3307
Verification
uv run --frozen pytest tests/client/test_streamable_http.py -q— 29 passedruff check/ruff format --checkclean;pyright0 errors;strict-no-coverpassed