fix(inference): adopt an interrupted managed vLLM container holding the serving port - #11530

Open

HOYALIM wants to merge 7 commits into

NVIDIA:mainfrom

HOYALIM:codex/11426-resume-adopt-leftover-vllm

Commits on Sep 11, 2026

fix(inference): adopt an interrupted managed vLLM container holding the serving port
HOYALIM
committed
f82b8d3
fix(inference): require the adopted container to hold the failed port
HOYALIM
committed
01be96c
fix(inference): keep adopted vLLM on local Docker
jyaunches
committed
52c4b37
merge: update main for validation
jyaunches
committed
452f254
fix(inference): match adopted vLLM binding address
jyaunches
committed
baed467
merge: update main for validation
jyaunches
committed
bf1b1d8
merge: update PR base for E2E qualification
jyaunches
committed
55b330e