Skip to content
This repository has been archived by the owner on Jan 11, 2023. It is now read-only.

add GPU ExtendedResourceToleration admission controller support #3181

Merged
merged 19 commits into from
Jun 27, 2018
Merged

add GPU ExtendedResourceToleration admission controller support #3181

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
19 commits
Select commit Hold shift + click to select a range
c5594e7
check for kubernetesConfig nil (#3164)
jackfrancis Jun 5, 2018
a36be0c
add GPU ExtendedResourceToleration admission controller support
lachie83 Jun 5, 2018
f39f29b
Revert "check for kubernetesConfig nil (#3164)"
lachie83 Jun 5, 2018
e9f061e
Update condition to match review
lachie83 Jun 5, 2018
2511e6f
Update conditionals to match both N series and device plugin
lachie83 Jun 6, 2018
c9e6dc1
Add toleration to run kube-proxy on tainted gpu nodes
lachie83 Jun 6, 2018
9e8fa64
Update Nvidia gpu device plugin to upstream ds example
lachie83 Jun 6, 2018
50b46a4
Update docs
lachie83 Jun 6, 2018
a8df245
Add support for 1.11
lachie83 Jun 7, 2018
83034c1
cleanup k8s_version_test
jackfrancis Jun 7, 2018
e91bc93
add nvidia.com/gpu to addons
jackfrancis Jun 7, 2018
bbc2c7a
Revert "add nvidia.com/gpu to addons"
lachie83 Jun 27, 2018
2ea4097
remove register node with taints
lachie83 Jun 27, 2018
6b7af96
Revert "Add toleration to run kube-proxy on tainted gpu nodes"
lachie83 Jun 27, 2018
b1b087a
Merge branch 'master' into feat-add-gpu-ExtendedResourceToleration
lachie83 Jun 27, 2018
478f39f
fixed rebase errors
lachie83 Jun 27, 2018
474d0e6
fix formatting errors
lachie83 Jun 27, 2018
cb1486e
fixed tests
lachie83 Jun 27, 2018
be6ebd7
actually check for nvidia addon enabled
jackfrancis Jun 27, 2018
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
8 changes: 4 additions & 4 deletions ...s-nvidia-gpu-device-plugin-daemonset.yaml → ...ddons-nvidia-device-plugin-daemonset.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,23 +2,23 @@ apiVersion: apps/v1
kind: DaemonSet
metadata:
labels:
k8s-app: nvidia-gpu-device-plugin
k8s-app: nvidia-device-plugin
kubernetes.io/cluster-service: "true"
addonmanager.kubernetes.io/mode: Reconcile
name: nvidia-gpu-device-plugin
name: nvidia-device-plugin
namespace: kube-system
spec:
selector:
matchLabels:
k8s-app: nvidia-gpu-device-plugin
k8s-app: nvidia-device-plugin
updateStrategy:
type: RollingUpdate
template:
metadata:
annotations:
scheduler.alpha.kubernetes.io/critical-pod: ""
labels:
k8s-app: nvidia-gpu-device-plugin
k8s-app: nvidia-device-plugin
spec:
priorityClassName: system-node-critical
affinity:
Expand Down
2 changes: 1 addition & 1 deletion parts/k8s/kubernetesmastercustomdata.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -281,7 +281,7 @@ MASTER_ARTIFACTS_CONFIG_PLACEHOLDER
{{end}}

{{if IsNVIDIADevicePluginEnabled}}
sed -i "s|<kubernetesNVIDIADevicePluginSpec>|{{WrapAsVariable "kubernetesNVIDIADevicePluginSpec"}}|g" "/etc/kubernetes/addons/nvidia-gpu-device-plugin.yaml"
sed -i "s|<kubernetesNVIDIADevicePluginSpec>|{{WrapAsVariable "kubernetesNVIDIADevicePluginSpec"}}|g" "/etc/kubernetes/addons/nvidia-device-plugin.yaml"
{{end}}

{{if EnableDataEncryptionAtRest }}
Expand Down
4 changes: 2 additions & 2 deletions pkg/acsengine/addons.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,8 +33,8 @@ func kubernetesAddonSettingsInit(profile *api.Properties) []kubernetesFeatureSet
true,
},
{
"kubernetesmasteraddons-nvidia-gpu-device-plugin-daemonset.yaml",
"nvidia-gpu-device-plugin.yaml",
"kubernetesmasteraddons-nvidia-device-plugin-daemonset.yaml",
"nvidia-device-plugin.yaml",
profile.IsNVIDIADevicePluginEnabled(),
},
{
Expand Down
90 changes: 46 additions & 44 deletions pkg/acsengine/k8s_versions.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,31 +9,32 @@ import (

var k8sComponentVersions = map[string]map[string]string{
"1.11": {
"dockerEngine": "1.13.*",
"dashboard": "kubernetes-dashboard-amd64:v1.8.3",
"exechealthz": "exechealthz-amd64:1.2",
"addon-resizer": "addon-resizer:1.8.1",
"heapster": "heapster-amd64:v1.5.1",
"metrics-server": "metrics-server-amd64:v0.2.1",
"kube-dns": "k8s-dns-kube-dns-amd64:1.14.8",
"addon-manager": "kube-addon-manager-amd64:v8.6",
"dnsmasq": "k8s-dns-dnsmasq-nanny-amd64:1.14.8",
"pause": "pause-amd64:3.1",
"tiller": "tiller:v2.8.1",
"rescheduler": "rescheduler:v0.3.1",
"aci-connector": "virtual-kubelet:latest",
"nodestatusfreq": DefaultKubernetesNodeStatusUpdateFrequency,
"nodegraceperiod": DefaultKubernetesCtrlMgrNodeMonitorGracePeriod,
"podeviction": DefaultKubernetesCtrlMgrPodEvictionTimeout,
"routeperiod": DefaultKubernetesCtrlMgrRouteReconciliationPeriod,
"backoffretries": strconv.Itoa(DefaultKubernetesCloudProviderBackoffRetries),
"backoffjitter": strconv.FormatFloat(DefaultKubernetesCloudProviderBackoffJitter, 'f', -1, 64),
"backoffduration": strconv.Itoa(DefaultKubernetesCloudProviderBackoffDuration),
"backoffexponent": strconv.FormatFloat(DefaultKubernetesCloudProviderBackoffExponent, 'f', -1, 64),
"ratelimitqps": strconv.FormatFloat(DefaultKubernetesCloudProviderRateLimitQPS, 'f', -1, 64),
"ratelimitbucket": strconv.Itoa(DefaultKubernetesCloudProviderRateLimitBucket),
"gchighthreshold": strconv.Itoa(DefaultKubernetesGCHighThreshold),
"gclowthreshold": strconv.Itoa(DefaultKubernetesGCLowThreshold),
"dockerEngine": "1.13.*",
"dashboard": "kubernetes-dashboard-amd64:v1.8.3",
"exechealthz": "exechealthz-amd64:1.2",
"addon-resizer": "addon-resizer:1.8.1",
"heapster": "heapster-amd64:v1.5.1",
"metrics-server": "metrics-server-amd64:v0.2.1",
"kube-dns": "k8s-dns-kube-dns-amd64:1.14.8",
"addon-manager": "kube-addon-manager-amd64:v8.6",
"dnsmasq": "k8s-dns-dnsmasq-nanny-amd64:1.14.8",
"pause": "pause-amd64:3.1",
"tiller": "tiller:v2.8.1",
"rescheduler": "rescheduler:v0.3.1",
"aci-connector": "virtual-kubelet:latest",
"nvidia-device-plugin": "k8s-device-plugin:1.10",
"nodestatusfreq": DefaultKubernetesNodeStatusUpdateFrequency,
"nodegraceperiod": DefaultKubernetesCtrlMgrNodeMonitorGracePeriod,
"podeviction": DefaultKubernetesCtrlMgrPodEvictionTimeout,
"routeperiod": DefaultKubernetesCtrlMgrRouteReconciliationPeriod,
"backoffretries": strconv.Itoa(DefaultKubernetesCloudProviderBackoffRetries),
"backoffjitter": strconv.FormatFloat(DefaultKubernetesCloudProviderBackoffJitter, 'f', -1, 64),
"backoffduration": strconv.Itoa(DefaultKubernetesCloudProviderBackoffDuration),
"backoffexponent": strconv.FormatFloat(DefaultKubernetesCloudProviderBackoffExponent, 'f', -1, 64),
"ratelimitqps": strconv.FormatFloat(DefaultKubernetesCloudProviderRateLimitQPS, 'f', -1, 64),
"ratelimitbucket": strconv.Itoa(DefaultKubernetesCloudProviderRateLimitBucket),
"gchighthreshold": strconv.Itoa(DefaultKubernetesGCHighThreshold),
"gclowthreshold": strconv.Itoa(DefaultKubernetesGCLowThreshold),
},
"1.10": {
"dockerEngine": "1.13.*",
Expand Down Expand Up @@ -243,25 +244,26 @@ func getK8sVersionComponents(version string, overrides map[string]string) map[st
"addonresizer": k8sComponentVersions["1.11"]["addon-resizer"],
"heapster": k8sComponentVersions["1.11"]["heapster"],
DefaultMetricsServerAddonName: k8sComponentVersions["1.11"]["metrics-server"],
"dns": k8sComponentVersions["1.11"]["kube-dns"],
"addonmanager": k8sComponentVersions["1.11"]["addon-manager"],
"dnsmasq": k8sComponentVersions["1.11"]["dnsmasq"],
"pause": k8sComponentVersions["1.11"]["pause"],
DefaultTillerAddonName: k8sComponentVersions["1.11"]["tiller"],
DefaultReschedulerAddonName: k8sComponentVersions["1.11"]["rescheduler"],
DefaultACIConnectorAddonName: k8sComponentVersions["1.11"]["aci-connector"],
"nodestatusfreq": k8sComponentVersions["1.11"]["nodestatusfreq"],
"nodegraceperiod": k8sComponentVersions["1.11"]["nodegraceperiod"],
"podeviction": k8sComponentVersions["1.11"]["podeviction"],
"routeperiod": k8sComponentVersions["1.11"]["routeperiod"],
"backoffretries": k8sComponentVersions["1.11"]["backoffretries"],
"backoffjitter": k8sComponentVersions["1.11"]["backoffjitter"],
"backoffduration": k8sComponentVersions["1.11"]["backoffduration"],
"backoffexponent": k8sComponentVersions["1.11"]["backoffexponent"],
"ratelimitqps": k8sComponentVersions["1.11"]["ratelimitqps"],
"ratelimitbucket": k8sComponentVersions["1.11"]["ratelimitbucket"],
"gchighthreshold": k8sComponentVersions["1.11"]["gchighthreshold"],
"gclowthreshold": k8sComponentVersions["1.11"]["gclowthreshold"],
"dns": k8sComponentVersions["1.11"]["kube-dns"],
"addonmanager": k8sComponentVersions["1.11"]["addon-manager"],
"dnsmasq": k8sComponentVersions["1.11"]["dnsmasq"],
"pause": k8sComponentVersions["1.11"]["pause"],
DefaultTillerAddonName: k8sComponentVersions["1.11"]["tiller"],
DefaultReschedulerAddonName: k8sComponentVersions["1.11"]["rescheduler"],
DefaultACIConnectorAddonName: k8sComponentVersions["1.11"]["aci-connector"],
"nodestatusfreq": k8sComponentVersions["1.11"]["nodestatusfreq"],
"nodegraceperiod": k8sComponentVersions["1.11"]["nodegraceperiod"],
"podeviction": k8sComponentVersions["1.11"]["podeviction"],
"routeperiod": k8sComponentVersions["1.11"]["routeperiod"],
"backoffretries": k8sComponentVersions["1.11"]["backoffretries"],
"backoffjitter": k8sComponentVersions["1.11"]["backoffjitter"],
"backoffduration": k8sComponentVersions["1.11"]["backoffduration"],
"backoffexponent": k8sComponentVersions["1.11"]["backoffexponent"],
"ratelimitqps": k8sComponentVersions["1.11"]["ratelimitqps"],
"ratelimitbucket": k8sComponentVersions["1.11"]["ratelimitbucket"],
"gchighthreshold": k8sComponentVersions["1.11"]["gchighthreshold"],
"gclowthreshold": k8sComponentVersions["1.11"]["gclowthreshold"],
DefaultNVIDIADevicePluginAddonName: k8sComponentVersions["1.11"]["nvidia-device-plugin"],
Copy link
Member

Choose a reason for hiding this comment

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

just curious, does NDP v1.10 work in k8s 1.11? when i tested NDP v1.9 with k8s 1.10 a while back, it didn't work

Copy link
Member Author

Choose a reason for hiding this comment

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

It appears to work the same way it does in 1.10

}
case "1.10":
ret = map[string]string{
Expand Down