Repository navigation
bug(client/streamable_http): prevent infinite SSE reconnect loops with bounded exponential backoff #3356
Closed as not planned
Description
Problem
In StreamableHTTPTransport.handle_get_stream, when an SSE stream disconnects or closes (server-side EOF/backpressure), attempt was reset to 0 upon exit. If an upstream server disconnects continuously, the client bypasses MAX_RECONNECTION_ATTEMPTS and enters an infinite reconnection loop without backoff jitter.
Expected Behavior
The client should respect MAX_RECONNECTION_ATTEMPTS on stream disconnects and apply exponential backoff to prevent thundering-herd storms on upstream MCP servers.
Activity
Ben-Home commentedon Aug 21, 2026
jagadeepmamidi commentedon Sep 1, 2026
I can reproduce the reconnect-loop behavior described here. I’d be interested in working on a focused fix that keeps the reconnection attempt count bounded across SSE disconnects, applies the existing backoff behavior, and adds regression coverage for repeated disconnects. Happy to work on this if the approach aligns with what you have in mind.
Kludex commentedon Oct 14, 2026
Duplicate of #3257.