Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in
Toggle navigation
Menu
Open sidebar
OpenDAS
dynamo
Commits
d9e8d334
Unverified
Commit
d9e8d334
authored
Oct 29, 2025
by
atchernych
Committed by
GitHub
Oct 29, 2025
Browse files
fix: adjust kgateway setup (#3963)
Signed-off-by:
Anna Tchernych
<
atchernych@nvidia.com
>
parent
3fe5653b
Changes
3
Hide whitespace changes
Inline
Side-by-side
Showing
3 changed files
with
10 additions
and
27 deletions
+10
-27
components/backends/vllm/deploy/agg.yaml
components/backends/vllm/deploy/agg.yaml
+5
-5
deploy/inference-gateway/README.md
deploy/inference-gateway/README.md
+0
-19
deploy/inference-gateway/install_gaie_crd_kgateway.sh
deploy/inference-gateway/install_gaie_crd_kgateway.sh
+5
-3
No files found.
components/backends/vllm/deploy/agg.yaml
View file @
d9e8d334
...
...
@@ -13,7 +13,7 @@ spec:
replicas
:
1
extraPodSpec
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
gitlab-master.nvidia.com:5005/dl/ai-dynamo/dynamo/dynamo:vllm-1029
VllmDecodeWorker
:
envFromSecret
:
hf-token-secret
dynamoNamespace
:
vllm-agg
...
...
@@ -24,12 +24,12 @@ spec:
gpu
:
"
1"
extraPodSpec
:
mainContainer
:
image
:
nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag
image
:
gitlab-master.nvidia.com:5005/dl/ai-dynamo/dynamo/dynamo:vllm-1029
workingDir
:
/workspace/components/backends/vllm
command
:
-
python3
-
-m
-
dynamo.vllm
-
python3
-
-m
-
dynamo.vllm
args
:
-
--model
-
Qwen/Qwen3-0.6B
deploy/inference-gateway/README.md
View file @
d9e8d334
...
...
@@ -56,25 +56,6 @@ INFERENCE_EXTENSION_VERSION=v0.5.1
kubectl apply
-f
https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/
$INFERENCE_EXTENSION_VERSION
/manifests.yaml
```
c. Install
`kgateway`
CRDs and kgateway.
```
bash
KGATEWAY_VERSION
=
v2.0.3
# Install the Kgateway CRDs
helm upgrade
-i
--create-namespace
--namespace
kgateway-system
--version
$KGATEWAY_VERSION
kgateway-crds oci://cr.kgateway.dev/kgateway-dev/charts/kgateway-crds
# Install Kgateway
helm upgrade
-i
--namespace
kgateway-system
--version
$KGATEWAY_VERSION
kgateway oci://cr.kgateway.dev/kgateway-dev/charts/kgateway
--set
inferenceExtension.enabled
=
true
```
d. Deploy the Gateway Instance
```
bash
kubectl create namespace my-model
kubectl apply
-f
https://github.com/kubernetes-sigs/gateway-api-inference-extension/raw/main/config/manifests/gateway/kgateway/gateway.yaml
-n
my-model
```
```
bash
kubectl get gateway inference-gateway
-n
my-model
...
...
deploy/inference-gateway/install_gaie_crd_kgateway.sh
View file @
d9e8d334
...
...
@@ -32,13 +32,15 @@ INFERENCE_EXTENSION_VERSION=v0.5.1
kubectl apply
-f
https://github.com/kubernetes-sigs/gateway-api-inference-extension/releases/download/
$INFERENCE_EXTENSION_VERSION
/manifests.yaml
-n
$MODEL_NAMESPACE
# Install
th
e Kgateway
CRDs and Kgateway
# Install
and upgrad
e Kgateway
(includes CRDs)
KGATEWAY_VERSION
=
v2.0.3
KGATEWAY_SYSTEM_NAMESPACE
=
kgateway-system
helm repo add kgateway-dev oci://cr.kgateway.dev/kgateway-dev
||
true
kubectl create namespace
$KGATEWAY_SYSTEM_NAMESPACE
||
true
helm upgrade
-i
--create-namespace
--namespace
$KGATEWAY_SYSTEM_NAMESPACE
--version
$KGATEWAY_VERSION
kgateway-crds oci://cr.kgateway.dev/kgateway-dev/charts/kgateway-crds
helm upgrade
-i
--namespace
$KGATEWAY_SYSTEM_NAMESPACE
--version
$KGATEWAY_VERSION
kgateway oci://cr.kgateway.dev/kgateway-dev/charts/kgateway
--set
inferenceExtension.enabled
=
true
# Deploy the Gateway Instance
kubectl apply
-f
https://github.com/kubernetes-sigs/gateway-api-inference-extension/raw/main/config/manifests/gateway/kgateway/gateway.yaml
-n
$MODEL_NAMESPACE
\ No newline at end of file
kubectl apply
-f
https://github.com/kubernetes-sigs/gateway-api-inference-extension/raw/v1.0.0/config/manifests/gateway/kgateway/gateway.yaml
-n
$MODEL_NAMESPACE
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment