Skip to content

Commit b4ddbd8

Browse files
spoltispolti
authored
fix typo on inferenceservice-config (kserve#4244)
Signed-off-by: Filippe Spolti <fspolti@redhat.com> Signed-off-by: Spolti <fspolti@redhat.com>
1 parent d20d422 commit b4ddbd8

2 files changed

Lines changed: 110 additions & 42 deletions

File tree

‎charts/kserve-resources/templates/configmap.yaml‎

Lines changed: 69 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,69 @@ data:
4242
"defaultImageVersion": "latest"
4343
}
4444
}
45+
# ====================================== ISVC CONFIGURATION ======================================
46+
# Example - setting custom annotation
47+
inferenceService: |-
48+
{
49+
"serviceAnnotationDisallowedList": [
50+
"my.custom.annotation/1"
51+
],
52+
"serviceLabelDisallowedList": [
53+
"my.custom.label.1"
54+
]
55+
}
56+
# Example - setting custom annotation
57+
inferenceService: |-
58+
{
59+
# ServiceAnnotationDisallowedList is a list of annotations that are not allowed to be propagated to Knative
60+
# revisions, which prevents the reconciliation loop to be triggered if the annotations is
61+
# configured here are used.
62+
# Default values are:
63+
# "autoscaling.knative.dev/min-scale",
64+
# "autoscaling.knative.dev/max-scale",
65+
# "internal.serving.kserve.io/storage-initializer-sourceuri",
66+
# "kubectl.kubernetes.io/last-applied-configuration"
67+
# Any new value will be appended to the list.
68+
"serviceAnnotationDisallowedList": [
69+
"my.custom.annotation/1"
70+
],
71+
# ServiceLabelDisallowedList is a list of labels that are not allowed to be propagated to Knative revisions
72+
# which prevents the reconciliation loop to be triggered if the labels is configured here are used.
73+
"serviceLabelDisallowedList": [
74+
"my.custom.label.1"
75+
]
76+
}
77+
# Example - setting custom resource
78+
inferenceService: |-
79+
{
80+
"resource": {
81+
"cpuLimit": "1",
82+
"memoryLimit": "2Gi",
83+
"cpuRequest": "1",
84+
"memoryRequest": "2Gi"
85+
}
86+
}
87+
# Example - setting custom resource
88+
inferenceService: |-
89+
{
90+
# resource contains the default resource configuration for the inference service.
91+
# you can override this configuration by specifying the resources in the inference service yaml.
92+
# If you want to unbound the resource (limits and requests), you can set the value to null or ""
93+
# or just remove the specific field from the config.
94+
"resource": {
95+
# cpuLimit is the limits.cpu to set for the inference service.
96+
"cpuLimit": "1",
97+
98+
# memoryLimit is the limits.memory to set for the inference service.
99+
"memoryLimit": "2Gi",
100+
101+
# cpuRequest is the requests.cpu to set for the inference service.
102+
"cpuRequest": "1",
45103
104+
# memoryRequest is the requests.memory to set for the inference service.
105+
"memoryRequest": "2Gi"
106+
}
107+
}
46108
# ====================================== STORAGE INITIALIZER CONFIGURATION ======================================
47109
# Example
48110
storageInitializer: |-
@@ -629,3 +691,10 @@ data:
629691
"memoryRequest": "{{ .Values.kserve.inferenceservice.resources.requests.memory }}"
630692
}
631693
}
694+
695+
opentelemetryCollector: |-
696+
{
697+
"scrapeInterval": "5s",
698+
"metricReceiverEndpoint": "keda-otel-scaler.keda.svc:4317",
699+
"metricScalerEndpoint": "keda-otel-scaler.keda.svc:4318"
700+
}

‎config/configmap/inferenceservice.yaml‎

Lines changed: 41 additions & 42 deletions
Original file line numberDiff line numberDiff line change
@@ -42,17 +42,17 @@ data:
4242
}
4343
}
4444
# ====================================== ISVC CONFIGURATION ======================================
45-
# Example
45+
# Example - setting custom annotation
4646
inferenceService: |-
47-
{
48-
"serviceAnnotationDisallowedList": [
49-
"my.custom.annotation/1"
50-
],
51-
"serviceLabelDisallowedList": [
52-
"my.custom.label.1"
53-
]
47+
{
48+
"serviceAnnotationDisallowedList": [
49+
"my.custom.annotation/1"
50+
],
51+
"serviceLabelDisallowedList": [
52+
"my.custom.label.1"
53+
]
5454
}
55-
# Example of isvc configuration
55+
# Example - setting custom annotation
5656
inferenceService: |-
5757
{
5858
# ServiceAnnotationDisallowedList is a list of annotations that are not allowed to be propagated to Knative
@@ -72,7 +72,38 @@ data:
7272
"serviceLabelDisallowedList": [
7373
"my.custom.label.1"
7474
]
75-
}
75+
}
76+
# Example - setting custom resource
77+
inferenceService: |-
78+
{
79+
"resource": {
80+
"cpuLimit": "1",
81+
"memoryLimit": "2Gi",
82+
"cpuRequest": "1",
83+
"memoryRequest": "2Gi"
84+
}
85+
}
86+
# Example - setting custom resource
87+
inferenceService: |-
88+
{
89+
# resource contains the default resource configuration for the inference service.
90+
# you can override this configuration by specifying the resources in the inference service yaml.
91+
# If you want to unbound the resource (limits and requests), you can set the value to null or ""
92+
# or just remove the specific field from the config.
93+
"resource": {
94+
# cpuLimit is the limits.cpu to set for the inference service.
95+
"cpuLimit": "1",
96+
97+
# memoryLimit is the limits.memory to set for the inference service.
98+
"memoryLimit": "2Gi",
99+
100+
# cpuRequest is the requests.cpu to set for the inference service.
101+
"cpuRequest": "1",
102+
103+
# memoryRequest is the requests.memory to set for the inference service.
104+
"memoryRequest": "2Gi"
105+
}
106+
}
76107
# ====================================== MultiNode CONFIGURATION ======================================
77108
# Example
78109
multiNode: |-
@@ -566,38 +597,6 @@ data:
566597
# This is to disable localmodel pv and pvc management for namespaces without isvcs
567598
"disableVolumeManagement": false
568599
}
569-
570-
# ====================================== LOCALMODEL CONFIGURATION ======================================
571-
# Example
572-
inferenceservice: |-
573-
{
574-
"resource": {
575-
"cpuLimit": "1",
576-
"memoryLimit": "2Gi",
577-
"cpuRequest": "1",
578-
"memoryRequest": "2Gi"
579-
}
580-
}
581-
inferenceservice: |-
582-
{
583-
# resource contains the default resource configuration for the inference service.
584-
# you can override this configuration by specifying the resources in the inference service yaml.
585-
# If you want to unbound the resource (limits and requests), you can set the value to null or ""
586-
# or just remove the specific field from the config.
587-
"resource": {
588-
# cpuLimit is the limits.cpu to set for the inference service.
589-
"cpuLimit": "1",
590-
591-
# memoryLimit is the limits.memory to set for the inference service.
592-
"memoryLimit": "2Gi",
593-
594-
# cpuRequest is the requests.cpu to set for the inference service.
595-
"cpuRequest": "1",
596-
597-
# memoryRequest is the requests.memory to set for the inference service.
598-
"memoryRequest": "2Gi"
599-
}
600-
}
601600
602601
explainers: |-
603602
{

0 commit comments

Comments
 (0)