k3s-agent job from k3s-wrapper/0.3.0
Github source:
643db8c
or
master branch
Properties¶
containerd_registry
¶
containerd registry configuration
k3s
¶
bind-address value
¶(listener) k3s bind address (default: 0.0.0.0)
- Default
0.0.0.0
bosh-post-start-delay-seconds
¶bosh post start tempo, to let the kubelet start the pods before bosh triggers another node update
- Default
30
containerd_additional_env_vars
¶additional env vars to set for containerd (the key will be prefixed with CONTAINERD_, and set in k3s launch context
- Default
[]
disable-vxlan-hardware-options
¶Disable VxLAN harware options on private interface
- Default
- tx-udp_tnl-segmentation - tx-udp_tnl-csum-segmentation
do-not-killall-on-post-stop
¶if set, the bosh post-stop script wont leverage k3s-killall.sh script
- Default
false
drain
¶
delete-emptydir-data
¶continue even if there are pods using emptyDir (local data that will be deleted when the node is drained).
- Default
true
disable-eviction
¶Force drain to use delete, even if eviction is supported. This will bypass checking PodDisruptionBudgets, use with caution
- Default
false
grace-period
¶period of time in seconds given to each pod to terminate gracefully. If negative, the default value specified in the pod will be used.
- Default
-1
ignore-daemonsets
¶Ignore DaemonSet-managed pods.
- Default
true
kubeconfig
¶bosh drain for agents need a k3s kubeconfig. If set, the drain will occur and use this kubeconfig
pod-selector
¶Label selector to filter pods on the node
selector
¶Selector (label query) to filter on
skip-wait-for-delete-timeout
¶If pod DeletionTimestamp older than N seconds, skip waiting for the pod. Seconds must be greater than 0 to skip.
- Default
0
timeout
¶The length of time to wait before giving up, zero means infinite
- Default
90s
kube-proxy-arg
¶(agent/flags) Customized flag for kube-proxy process
- Default
[]
kubelet-args
¶(agent/flags) Customized flag for kubelet process
- Default
[]
kubelet-config-file
¶content of kubelet config file, to enable eg GraceFull Node Shutdown
- Default
|+ apiVersion: kubelet.config.k8s.io/v1beta1 kind: KubeletConfiguration shutdownGracePeriod: 30s shutdownGracePeriodCriticalPods: 10s
master_vip_api
¶externaly defined vip ip for HA k3s (enables multi master instance groups). This is used for public api access, tls-san, and agents to server communication
node-labels
¶(agent/node) Registering and starting kubelet with set of labels format is key=value
- Default
[]
node-taints
¶(agent/node) Registering kubelet with set of taints.format is key=value:Effect)
- Default
[]
node_name_prefix
¶explicitly set k8s node name. If not set, - is set automatically. If set, name is -
set-provider-id-prefix
¶If set, the default provider id (k3s://- will be set as kubelet arg as ://-
token
¶(cluster) Shared secret used to join a server or agent to a cluster [$K3S_TOKEN]
token-file-content
¶token-file content. see https://kubernetes.io/docs/reference/access-authn-authz/authentication/#static-token-file
v
¶(logging) Number for the log level verbosity (default: 0)
- Default
0
registry
¶
mirrors
¶
tls
¶
ca
¶private registry ca
cert
¶private registry certificate
key
¶private registry private key
Templates¶
Templates are rendered and placed onto corresponding
instances during the deployment process. This job's templates
will be placed into /var/vcap/jobs/k3s-agent/
directory
(learn more).
bin/ctl
(frombin/ctl.erb
)bin/drain
(frombin/drain.erb
)bin/envrc
(frombin/envrc
)bin/k3s-killall.sh
(frombin/k3s-killall.sh
)bin/post-deploy
(frombin/post-deploy.erb
)bin/post-start
(frombin/post-start.erb
)bin/post-stop
(frombin/post-stop.erb
)bin/pre-start
(frombin/pre-start.erb
)bin/pre-stop
(frombin/pre-stop.erb
)bin/setup-user-env
(frombin/setup-user-env.erb
)config/bpm.yml
(fromconfig/bpm.yml
)config/kubelet-config.yaml
(fromconfig/kubelet-config.yaml.erb
)config/registries.yaml
(fromconfig/registries.yaml.erb
)config/registry.ca
(fromconfig/registry.ca.erb
)config/registry.cert
(fromconfig/registry.cert.erb
)config/registry.key
(fromconfig/registry.key.erb
)config/token.csv
(fromconfig/token.csv.erb
)
Packages¶
Packages are compiled and placed onto corresponding
instances during the deployment process. Packages will be
placed into /var/vcap/packages/
directory.
This job relies on no runtime packages.