Skip to content

Commit 2f84005

Browse files
authored
fix-enhance-install-doc (llm-d#828)
1 parent 3557902 commit 2f84005

1 file changed

Lines changed: 11 additions & 3 deletions

File tree

charts/workload-variant-autoscaler/README.md

Lines changed: 11 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -64,7 +64,7 @@ helm upgrade -i workload-variant-autoscaler ./workload-variant-autoscaler \
6464
After a WVA controller has been installed,
6565
you can add one or more models running in LLMD namespaces as scale targets to the WVA controller. As an example, the following command adds model name `my-model-a` with model ID `meta-llama/Llama-3.1-8` running in `team-a` LLMD namespace. This command creates the corresponding VA, HPA resources in `team-a` namespace.
6666
```
67-
helm install -i wva-model-a ./workload-variant-autoscaler \
67+
helm install wva-model-a ./workload-variant-autoscaler \
6868
-n $WVA_NS \
6969
--set controller.enabled=false \
7070
--set va.enabled=true \
@@ -75,7 +75,7 @@ helm install -i wva-model-a ./workload-variant-autoscaler \
7575
```
7676
Here is an example to add another model to the same WVA controller:
7777
```
78-
helm install -i wva-model-b ./workload-variant-autoscaler \
78+
helm install wva-model-b ./workload-variant-autoscaler \
7979
-n $WVA_NS \
8080
--set controller.enabled=false \
8181
--set va.enabled=true \
@@ -84,7 +84,15 @@ helm install -i wva-model-b ./workload-variant-autoscaler \
8484
--set llmd.modelName=my-model-b \
8585
--set llmd.modelID="Qwen/Qwen3-0.6B"
8686
```
87-
87+
**Notes**:
88+
- When there are multiple WVA controllers installed in different namespaces, there's the possibility of adding models in a LLMD namespace as scale targets using the **same** `release name`. If `helm install` was used to add then there will be a clear message such as:
89+
```
90+
INSTALLATION FAILED: cannot re-use a name that is still in use
91+
```
92+
However, if `helm upgrade -i` (combine upgrade and install) was used then the message is less clear as shown below. In this case, different release names should be used:
93+
```
94+
Error: UPGRADE FAILED: Unable to continue with update: Service "workload-variant-autoscaler-vllm" in namespace "xyz" exists and cannot be imported into the current release: invalid ownership metadata; annotation validation error: key "meta.helm.sh/release-namespace" must equal "abc": current value is "xyz"
95+
```
8896

8997
## Values
9098

0 commit comments

Comments
 (0)