[FIX]: resolve server_routes_file pointer in gumcp transport config - #33
Merged
Merged
Conversation
server_routes now ship as a sandbox file instead of inline in the GUMCP_CONFIG env var (oversized env blocks broke every sandbox exec with E2BIG for orgs with many gumstack servers). the transport loads the file into the client config and keys the cached client on the file's mtime so a refreshed per-call file rebuilds the session. Co-authored-by: Cursor <cursoragent@cursor.com>
waiho-gumloop
approved these changes
Jul 24, 2026
dvlpjrs
reviewed
Jul 24, 2026
dvlpjrs
reviewed
Jul 24, 2026
dvlpjrs
reviewed
Jul 24, 2026
dvlpjrs
approved these changes
Jul 24, 2026
dvlpjrs
left a comment
Collaborator
There was a problem hiding this comment.
Approved: you might need to upgrade version
An unreadable or corrupt routes file left routed servers silently missing from the config. Keep the degrade, since a client with only gateway servers has no routes file at all, but say so at warning level. Drops two comments that restated the code. Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Collaborator
Author
|
Bumped to 0.4.6 in 4e3cf86 ( |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
server_routesare moving out of theGUMCP_CONFIGenv var into a sandbox file (server_routes_filepointer key), because inline routes grow ~1.3KB per visible Gumstack server and blow past the kernel's exec arg/env limit — every process spawn in affected sandboxes fails withfork/exec: argument list too long(E2BIG). Production incident: interactionubHTArDjKdRPXwBjRdCw72, 1.8k+ log hits over 72h for Gumloop-org users with 60+ visible servers._load_config()now resolves theserver_routes_filepointer into inlineserver_routesbefore constructing thegumcp_client.AsyncClient(tolerates missing/malformed files — non-routed servers keep working).GUMCP_CONFIGstring does.Backward/forward compatible: with today's backend (inline routes, no pointer) this is a no-op. The backend change that writes the file and slims the env var lands separately in
gumloop_backend; a companiongumcp-clientrelease adds the same pointer resolution for scripts that construct clients directly.Test plan
uv run pytest tests/sdk/test_gumcp_transport.py— 24 passed, including new tests for pointer resolution, missing/malformed routes file tolerance, and fingerprint invalidation on routes-file mtime changeuv run ruff checkclean on both filesMade with Cursor