| FazBrowse GitHub Viewer | Trending | | Home |
| Tools: [Download Repo ZIP] [Original HTTPS Page] |
- Add per-server mutex to MCP to prevent concurrent state mutations - Clean up transports on connection failure and for unused connections - Add connection/process cleanup on LSP client init failure - Add proactive stale subscriber cleanup and onError handler in PTY - Replace silent catches with debug logging in config, watcher, instance These fixes address potential race conditions, resource leaks, and improve debuggability by logging previously swallowed errors.
|
Thanks for your contribution! This PR doesn't have a linked issue. All PRs must reference an existing issue. Please:
See CONTRIBUTING.md for details. |
Sorry, something went wrong.
|
The following comment was made by an LLM, it may be inaccurate: The search results show the current PR (#9047) and two other PRs related to memory leak prevention: Potential Related PRs:
These PRs address similar concerns around cleanup and disposal of resources, which aligns with the current PR's work on LSP client cleanup and transport cleanup. However, they appear to be older PRs focused on subscription memory leaks rather than the comprehensive async operation reliability improvements in PR #9047. The current PR #9047 appears to be a newer, more comprehensive fix that builds on or complements these earlier cleanup efforts. |
Sorry, something went wrong.
| Back | FazBrowse Home | New Git URL |
Summary
This PR adds several reliability improvements for async operations in the codebase.
Changes
MCP State Synchronization (mcp/index.ts)
LSP Client Cleanup (lsp/client.ts)
PTY Subscriber Management (pty/index.ts)
Improved Error Logging
Testing
These changes have been tested on the agent-core fork and all 721 tests pass.
Motivation
These fixes address: