fix: Model readiness for namespaced models#502
Merged
yinggeh merged 1 commit intoJun 10, 2026
Conversation
yinggeh
changed the title
Strict server readiness and ModelIsReady resolved models by name only, which is ambiguous when model namespacing is enabled and the same model name exists in multiple namespaces, causing L0_model_namespacing to fail. IsReady now resolves models via the full ModelIdentifier, and ModelIsReady takes an already-resolved Model so each caller controls model resolution.
yinggeh
force-pushed
the
yinggeh/tri-1341-fix-ci-test-l0_model_namespacing-base
branch
from
7de0d9f to
632ad53
Compare
yinggeh
changed the title
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does the PR do?
Fixes the
L0_model_namespacing--baseandL0_metrics--basetests. Strict server readiness (InferenceServer::IsReady) andInferenceServer::ModelIsReadyresolved models by name only, which is ambiguous when model namespacing is enabled and the same model name exists in multiple namespaces — the lookup fails and the server is reported not-ready. Readiness now resolves models by the fullModelIdentifier.This regression was introduced by #473 (
fix: /v2/health/ready returns 200 when Python backend stub is dead (#8604), commit 53fc26e), which added the runtime backend readiness check in the strict-readiness path using name-only model lookup.Checklist
<commit_type>: <Title>Commit Type:
Related PRs:
Where should the reviewer start?
src/server.cc/src/server.h:IsReadystrict-readiness loop now resolves each model via the fullModelIdentifier(GetModel(mv.first, ...));ModelIsReadytakes an already-resolvedModel&.src/tritonserver.cc:TRITONSERVER_ServerModelIsReadyresolves the model (name-based, matching the public C-API contract) and calls the newModelIsReadyoverload.Test plan:
L0_model_namespacing--baseL0_metrics--baseCI Pipeline ID: 53863861
Caveats:
The public C-API
TRITONSERVER_ServerModelIsReadystill resolves by name (no namespace parameter in the API contract); namespaced readiness resolution applies to the internal strict-readiness path.Background
Related Issues:
N/A