# NousResearch/hermes-agent · MCP OAuth and HTTP transport reconnect deadlock fixes

## Open now

- 1 issue, 1 PR open on GitHub.
- #31987: MCP HTTP transport: anyio RuntimeError during streamable_http_client cleanup causes reconnect failure loop
- PR #38198: fix: Close inner SDK auth generator to prevent OAuth reconnect deadlock

These issues and fixes address deadlocks and reconnect failures in MCP OAuth-backed servers, specifically around stranded auth-flow generator locks and anyio cleanup errors during streamable HTTP transport teardown. Multiple fix attempts and salvage PRs suggest ongoing iteration on the same underlying concurrency problems.

## TL;DR

- 10 member(s): 1 open issue(s), 2 closed issue(s), 1 open PR(s), 1 merged PR(s), 5 PR(s) closed without merging.
- A fix was merged to main in PR #97458. Verify it addresses your case.
- **Recommended:** read PR #97458 (and any competing PRs below) before filing. Confirm the merged change covers your symptom.

## Maintainer resolution verdict

- Internal-tracker verdict over this cluster's issue members: `unresolved`. This is distinct from the fix status above. Weigh it yourself.

## How this cluster was grouped

- Link evidence (`STRUCTURAL_PARTIAL`): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
- Members outside that structural core may belong to a different underlying problem.
- Pairwise link density `0.17`: the share of member pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. `1.00` means every pair is linked.

## Open related issues

- #31987: MCP HTTP transport: anyio RuntimeError during streamable_http_client cleanup causes reconnect failure loop

## Closed / duplicate issues

- #38193: \[Bug\]: OAuth-backed MCP server permanently deadlocks after keepalive reconnect: auth-flow generator's lock is released cross-task.
- #49543: OAuth MCP servers (e.g. Honeycomb) drop from active toolset mid-session — RuntimeError: The current task is not holding this lock + 120s call hangs

## Prior & competing PRs

- **#97458: fix(mcp-oauth): avoid locking across resource requests**. merged to main · fix tier `verified`
  - maintainer direction: category `merged`
  - files: `tests/tools/test_mcp_oauth_bidirectional.py`, `tests/tools/test_mcp_oauth_manager.py`, `tools/mcp_oauth_manager.py`
- **#32111: fix(tools): ignore anyio cleanup artifact during MCP HTTP reconnect**. closed without merge
  - maintainer direction: category `stale_abandoned`; implication `active`
- **#38198: fix: Close inner SDK auth generator to prevent OAuth reconnect deadlock**. open
  - maintainer direction: category `uncategorized`; implication `needs_llm`; refs `#63495`
  - files: `tests/tools/test_mcp_oauth_generator_cleanup.py`, `tools/mcp_oauth_manager.py`
- **#63495: fix(mcp-oauth): close inner SDK auth generator on teardown (salvage #38198)**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#97458`, `#90888`
- **#76526: fix: close stranded MCP OAuth auth flows**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#97458`, `#91049`
- **#90888: fix(mcp-oauth): bind delegated auth-flow teardown to the owning task**. closed without merge
  - maintainer direction: category `implemented_elsewhere`; implication `superseded`; refs `7c6f5ce3b717ccbb4b867bbb66855fdc8ab5dd9b`, `b2c4f1f376167e7e34a88c3dbd544e1fdc848c14`, `#38193`, `#84963`, `#88875`
- **#91049: fix(mcp-oauth): recover stranded auth-flow locks**. closed without merge
  - maintainer direction: category `superseded_by_pr`; implication `superseded`; refs `#97458`

## Fix status

- Merged to main in PR #97458 (cited above). No release data is tracked. Verify downstream.
- Fix last verified (merged) `2026-08-29`.
- Subsystem: `mcp`