Skip to content

vllm: support to run as non-root #882

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 19, 2025
Merged

vllm: support to run as non-root #882

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 0 additions & 3 deletions helm-charts/agentqna/gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,6 @@ vllm:
VLLM_SKIP_WARMUP: true
shmSize: 16Gi
extraCmdArgs: ["--tensor-parallel-size", "4", "--max-seq_len-to-capture", "16384", "--enable-auto-tool-choice", "--tool-call-parser", "llama3_json"]
securityContext:
runAsUser: 0
runAsNonRoot: false

supervisor:
llm_endpoint_url: http://{{ .Release.Name }}-vllm
Expand Down
10 changes: 0 additions & 10 deletions helm-charts/chatqna/gaudi-vllm-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,18 +15,8 @@ vllm:
resources:
limits:
habana.ai/gaudi: 1
securityContext:
runAsUser: 0
runAsNonRoot: false
startupProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1
failureThreshold: 360
readinessProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1

PT_HPU_ENABLE_LAZY_COLLECTIVES: "true"
OMPI_MCA_btl_vader_single_copy_mechanism: "none"
Expand Down
10 changes: 0 additions & 10 deletions helm-charts/chatqna/guardrails-gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,18 +87,8 @@ vllm:
resources:
limits:
habana.ai/gaudi: 1
securityContext:
runAsUser: 0
runAsNonRoot: false
startupProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1
failureThreshold: 360
readinessProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1

PT_HPU_ENABLE_LAZY_COLLECTIVES: "true"
OMPI_MCA_btl_vader_single_copy_mechanism: "none"
Expand Down
3 changes: 0 additions & 3 deletions helm-charts/common/agent/gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,5 @@ vllm:
PT_HPU_ENABLE_LAZY_COLLECTIVES: true
VLLM_SKIP_WARMUP: true
extraCmdArgs: ["--tensor-parallel-size", "4", "--max-seq_len-to-capture", "16384", "--enable-auto-tool-choice", "--tool-call-parser", "llama3_json"]
securityContext:
runAsUser: 0
runAsNonRoot: false

llm_endpoint_url: http://{{ .Release.Name }}-vllm
3 changes: 0 additions & 3 deletions helm-charts/common/llm-uservice/vllm-docsum-gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,8 +25,5 @@ vllm:
resources:
limits:
habana.ai/gaudi: 1
securityContext:
runAsUser: 0
runAsNonRoot: false
startupProbe:
failureThreshold: 360
3 changes: 0 additions & 3 deletions helm-charts/common/llm-uservice/vllm-gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,6 @@ vllm:
resources:
limits:
habana.ai/gaudi: 1
securityContext:
runAsUser: 0
runAsNonRoot: false
startupProbe:
failureThreshold: 360

Expand Down
6 changes: 1 addition & 5 deletions helm-charts/common/vllm/gaudi-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,10 +13,6 @@ resources:
limits:
habana.ai/gaudi: 1

# NOTE: opea/vllm-gaudi image requires running as root during runtime
securityContext:
runAsUser: 0
runAsNonRoot: false
# NOTE: opea/vllm-gaudi needsd more warm up time
# NOTE: opea/vllm-gaudi needs more warm up time
startupProbe:
failureThreshold: 360
1 change: 1 addition & 0 deletions helm-charts/common/vllm/templates/configmap.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,7 @@ data:
NUMBA_CACHE_DIR: "/tmp"
HF_HOME: "/tmp/.cache/huggingface"
XDG_CONFIG_HOME: "/tmp"
TORCHINDUCTOR_CACHE_DIR: "/tmp/pytorchinductor_cache"
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You could add here a comment such as "PyTorch requires root user if cache files are not owned by user".

# https://github.com/outlines-dev/outlines/blob/main/outlines/caching.py#L14-L29
OUTLINES_CACHE_DIR: "/tmp/.cache/outlines"
{{- if .Values.VLLM_CPU_KVCACHE_SPACE }}
Expand Down
10 changes: 0 additions & 10 deletions helm-charts/docsum/gaudi-vllm-values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,18 +19,8 @@ vllm:
resources:
limits:
habana.ai/gaudi: 1
securityContext:
runAsUser: 0
runAsNonRoot: false
startupProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1
failureThreshold: 360
readinessProbe:
initialDelaySeconds: 5
periodSeconds: 5
timeoutSeconds: 1

PT_HPU_ENABLE_LAZY_COLLECTIVES: "true"
OMPI_MCA_btl_vader_single_copy_mechanism: "none"
Expand Down