Standardize container/service names with functional prefixes#32
Merged
Conversation
Evrard-Nil
approved these changes
May 11, 2026
Evrard-Nil
force-pushed
the
main
branch
2 times, most recently
from
8c48268 to
b19e44a
Compare
- model-vllm-*: vLLM inference engines (DeepSeek, GLM-5 staying as model-sg-* since SGLang, GLM-5.1 same, Qwen3.5, Qwen3-30B, GPT-OSS, gpt-oss-single, Qwen3-VL, Qwen3-Embedding, Qwen3-Reranker, Whisper, Gemma-4-31B) - model-sg-*: SGLang inference engines (GLM-5, GLM-5.1, FLUX, Qwen3.6-35B) - model-privacy-filter: custom HF inference service - proxy-*: inference-proxy (vllm-proxy-rs) and nginx reverse proxy All internal hostname references (VLLM_BASE_URL, VLLM_BACKEND_URLS, openmetrics endpoints, registrar health checks, nginx proxy_pass) updated to match. FLUX Datadog labels switched from vllm to sglang/openmetrics since FLUX runs SGLang despite its previous vllm- prefix.
lloydmak99
force-pushed
the
feat/standardize-container-names
branch
from
5c66217 to
abb3c49
Compare
added 2 commits
May 26, 2026 19:35
Merge origin/main into feat/standardize-container-names to resolve merge conflicts. Resolve modify/delete conflicts: - DeepSeek-V3.1.yaml, GLM-5.yaml, gpt-oss-single.yaml: deleted on main (consolidated into other files); keep deletion. Resolve content conflicts by starting from main's version and re-applying the PR's naming conventions: - nginx → proxy-nginx - vllm-proxy-* → proxy-* - glm51 → model-sg-glm51 - qwen35-1/2 → model-sg-qwen35-122b-1/2 (now SGLang on main) - vllm-qwen3-30b → model-vllm-qwen3-30b - vllm-gpt-oss-* → model-vllm-gpt-oss-* - vllm-qwen3-6-35b-* → model-sg-qwen3-6-35b-* - vllm-gemma-4-31b → model-sg-gemma-4-31b (now SGLang on main) - vllm-flux-2-klein-4b-1 → model-sg-flux-2-klein-4b-1 - vllm-qwen3-vl → model-vllm-qwen3-vl - vllm-qwen3-embeddings → model-vllm-qwen3-embeddings - vllm-qwen3-reranker → model-vllm-qwen3-reranker - vllm-whisper3-large → model-vllm-whisper3-large - privacy-filter → model-privacy-filter All internal references updated: VLLM_BASE_URL, VLLM_BACKEND_URLS, openmetrics endpoints, registrar health checks, nginx proxy_pass directives, and Datadog service labels. Also: add sglang:* metric filter to FLUX Datadog openmetrics config (matching what Qwen3.6 and Gemma-4 already use).
The anchor &qwen35-122b-common was not updated to &sg-qwen35-122b-common when the YAML key was renamed, causing 'unknown anchor' validation errors.
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.
Summary
Rebased onto current main; covers all model files including the Qwen3.6-35B-A3B-FP8 and Gemma-4-31B-it services that were added after the original PR.
Convention:
model-vllm-*— vLLM inference engines (DeepSeek-V3.1, Qwen3.5-122B, Qwen3-30B, GPT-OSS x2, Qwen3-VL, Qwen3-Embedding, Qwen3-Reranker, Whisper, Gemma-4-31B, gpt-oss-single)model-sg-*— SGLang inference engines (GLM-5, GLM-5.1, FLUX, Qwen3.6-35B x2); also corrects FLUX which was previously misprefixedvllm-despite running SGLangmodel-privacy-filter— custom HF inference service (no engine prefix)proxy-nginx— nginx reverse proxyproxy-*— all inference-proxy (vllm-proxy-rs) instancesAll internal references updated throughout:
VLLM_BASE_URL,VLLM_BACKEND_URLS, openmetrics endpoints, registrar health checks, and nginxproxy_passdirectives. FLUX Datadog labels switched fromvllm→sglang/openmetricswithsglang:*metric filter (matches what Qwen3.6 already does on main).Test plan
docker compose upresolves all service hostnames correctly after renamesglang)