docs: record the Fabric agent-eval false-failure known issue - #1261
Open
SandyChapman wants to merge 3 commits into
Open
docs: record the Fabric agent-eval false-failure known issue#1261SandyChapman wants to merge 3 commits into
SandyChapman wants to merge 3 commits into
Conversation
A Fabric agent-eval trial can be scored as an adapter failure after the agent completed its work, because NeMo Relay's LangChain callback handler leaves a scope open when concurrent LangGraph sibling runs finish out of LIFO order. The defect is fixed upstream in NeMo-Relay #755 but cannot reach the platform until NeMo Relay publishes a release carrying it and NeMo Fabric moves off its nemo-relay<0.7 pin, so document it as a known issue for this release. Signed-off-by: Sandy Chapman <schapman@nvidia.com>
15 tasks
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Enterprise Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
📝 WalkthroughWalkthroughThe release notes add a Known issues section. It documents that Deep Agents trials using the NeMo Fabric adapter can receive failure scores after successful agent completion. A future release fixes the issue. ChangesFabric evaluation documentation
Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
Signed-off-by: Sandy Chapman <schapman@nvidia.com>
Contributor
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/about/release-notes/current-release.mdx`:
- Around line 210-212: Update the “Deep Agents trials can be scored as failures”
release-note entry to explicitly state that no workaround is available in the
current release until the fix reaches a future release.
🪄 Autofix
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: cc208651-7b33-4eb2-b04b-cd9670f4cb3c
📒 Files selected for processing (1)
docs/about/release-notes/current-release.mdx
Signed-off-by: Sandy Chapman <schapman@nvidia.com>
Contributor
|
SandyChapman
enabled auto-merge
August 12, 2026 19:41
github-merge-queue
Bot
removed this pull request from the merge queue due to failed status checks
Aug 12, 2026
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
2 participants
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.
Overview
What. Documents the Fabric agent-eval false-failure as a known issue in the current release notes.
Why. A trial run through a NeMo Fabric adapter can be scored as an adapter failure even though the agent completed its work and produced the expected output, so evaluation results can understate an agent's accuracy. Users have no workaround in this release, and the fix cannot ship here yet, so the behaviour needs to be stated rather than left to be rediscovered.
Details
The root cause is upstream, not in the evaluator: NeMo Relay closes telemetry scopes strictly LIFO, but LangGraph schedules concurrent sibling chain runs that finish in an order the stack rejects. Relay's LangChain callback handler dropped the rejected close and left the scope live, so the enclosing request scope failed with
scope handle is not at the top of the stack, and the Fabric adapter reported that teardown error as an invocation failure. It reproduces at any parallelism, including one.Fixed upstream in NeMo-Relay #755, merged to
release/0.7. It cannot reach NeMo Platform yet: Relay has not cut a release containing it, and NeMo Fabric pinsnemo-relay>=0.6.0,<0.7, which excludes the whole 0.7 line. So the chain is a Relay 0.7.x release, then a Fabric dependency update, then a Fabric rev bump here.This supersedes #1205, an evaluator-side mitigation that treated the teardown error as non-fatal. That worked around the symptom rather than fixing it, so it is closed in favour of the upstream fix plus this disclosure.
Placed under a new
## Known issuesheading rather than in## Current constraints, since that section lists intentional scope limits and this is a defect.Testing. Docs-only change; no code paths affected.
pre-commitpasses on the changed file.Breaking changes. None.
Where should the reviewer start?
docs/about/release-notes/current-release.mdx— the new## Known issuessection. Worth checking the impact wording is right for a public docs surface.Related Issues:
Summary by CodeRabbit