Repository navigation
feat: show full alias groups in cluster status output - #8
Merged
Merged
Conversation
Aliases sharing one runtime model (unified layout) collapsed to the first alias in status display: model lists, tf_user attribution, and disk totals were misleading. normalize_model_statuses now exposes the alias group; plain and rich status show 'agent,coder,memory (runtime-id)', attribute active edge requests across the whole group, and count shared runtime models once in disk sums. Decouple inference-health test from repo tfconfig disk sizes.
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars. ⚙️ Run configuration
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
With the unified node layout several public aliases (
agent,coder,memory) share oneruntime_model_id. Status output collapsed each runtime model to its first alias, which is misleading:agentwhilecoder/memoryrequests were being served by the same loaded model;tf_userattribution matched only the first alias, so an activecoderrequest showedtf_user=-;hot_loaded_models (318.9G)for one 106.3 GB model).Changes
normalize_model_statusesnow emitsaliases: [...]for every runtime model (contract addition, backward compatible).agent,coder,memory (Qwen3.8-Flash-Next-oQ4e-mtp)when an alias group shares a runtime model, attribute active edge requests across the whole group, and dedupe disk sums byruntime_model_id.aliasesfall back to the previous single-alias shape, so CLI and edge can be upgraded in any order.test_cluster_status_reports_inference_healthfrom repotfconfig.yamldisk sizes (it silently broke whenever the registry changed).Evidence
TDD: four RED→GREEN cycles (alias group in normalize, plain display + group-wide attribution, disk dedupe, rich table). Full suite: 403 passed (3 new tests, 1 updated),
ruff checkclean. Live check of the disk dedupe against the real cluster:served_models (109.5G)/hot_loaded_models (106.3G)instead of 318.9G. Alias-group display goes live on nodes after the edge service redeploys this code.