Gardener uses the machine API and leverages the functionalities of the machine-controller-manager (MCM) in order to manage the worker nodes of a shoot cluster. The machine-controller-manager itself simply takes a reference to an OS-image and (optionally) some user-data (a script or configuration that is executed when a VM is bootstrapped), and forwards both to the provider's API when creating VMs. MCM does not have any restrictions regarding supported operating systems as it does not modify or influence the machine's configuration in any way - it just creates/deletes machines with the provided metadata.
Consequently, Gardener needs to provide this information when interacting with the machine-controller-manager. This means that basically every operating system is possible to be used as long as there is some implementation that generates the OS-specific configuration in order to provision/bootstrap the machines.
- The operating system must have built-in Docker support.
- The operating system must have systemd support.
- The operating system must have
wgetpre-installed. - The operating system must have
jqpre-installed.
The reasons for that will become evident later.
Gardener installs a few components onto every worker machine in order to allow it to join the shoot cluster.
There is the kubelet process, some scripts for continuously checking the health of kubelet and docker, but also configuration for log rotation, CA certificates, etc.
The complete configuration you can find here. We are calling this the "original" user-data.
Usually, you would submit all the components you want to install onto the machine as part of the user-data during creation time. However, some providers do have a size limitation (like ~16KB) for that user-data. That's why we do not send the "original" user-data to the machine-controller-manager (who forwards it then to the provider's API). Instead, we only send a small script that downloads the "original" data and applies it on the machine directly. This way we can extend the "original" user-data without any size restrictions - plus we can modify it without the necessity of re-creating the machine (because we run a script that downloads and updates it continuously).
The high-level flow is as follows:
-
For every worker pool
Xin theShootspecification, Gardener creates aSecretnamedcloud-config-<X>in thekube-systemnamespace of the shoot cluster. The secret contains the "original" user-data. -
Gardener generates a kubeconfig with minimal permissions just allowing reading these secrets. It is used by the
downloaderscript later. -
Gardener provides the
downloaderscript, the kubeconfig, and the machine image stated in theShootspecification to the machine-controller-manager. -
Based on this information the machine-controller-manager creates the VM.
-
After the VM has been provisioned the
downloaderscript starts and fetches the appropriateSecretfor its worker pool (containing the "original" user-data) and applies it.
With gardener v1.23 a file with the content <<BOOTSTRAP_TOKEN>> is added to the cloud-config-<worker-group>-downloader OperatingSystemConfig (part of step 2 in the graphic below).
Via the OS extension the new file (with its content in clear-text) gets passed to the corresponding Worker resource.
The Worker controller has to guarantee that:
- a bootstrap token is created.
- the
<<BOOTSTRAP_TOKEN>>in the user data is replaced by the generated token. One implementation of that is depicted in the picture where the machine-controller-manager creates a temporary token and replaces the placeholder.
As part of the user-data the bootstrap-token is placed on the newly created VM under a defined path. The cloud-config-script will then refer to the file path of the added bootstrap token in the kubelet-bootstrap script.
With Gardener v1.23, we replaced the long-valid bootstrap-token shared between nodes with a short-lived token unique for each node, ref: #3898.
❗ When updating to Gardener version >=1.35 the old bootstrap-token will be removed. You are required to update your extensions to the following versions when updating Gardener:
| Extension | Version | Release Date | Pull Request |
|---|---|---|---|
| os-gardenlinux | v0.9.0 | 2 Jul | gardener/gardener-extension-os-gardenlinux#29 |
| os-suse-chost | v1.11.0 | 2 Jul | gardener/gardener-extension-os-suse-chost#41 |
| os-ubuntu | v1.11.0 | 2 Jul | gardener/gardener-extension-os-ubuntu#42 |
| os-flatcar | v1.7.0 | 2 Jul | gardener/gardener-extension-os-coreos#24 |
| infrastructure-provider using Machine Controller Manager | varies | ~ end of 2019 | gardener/machine-controller-manager#351 |
With ongoing development and new releases of Gardener some new components could be required to get installed onto every shoot worker VM, or existing components need to be changed.
Gardener achieves that by simply updating the user-data inside the Secrets mentioned above (step 1).
The downloader script is continuously (every 30s) reading the secret's content (which might include an updated user-data) and storing it onto the disk.
In order to re-apply the (new) downloaded data the secrets do not only contain the "original" user-data but also another short script (called "execution" script).
This script checks whether the downloaded user-data differs from the one previously applied - and if required - re-applies it.
After that it uses systemctl to restart the installed systemd units.
With the help of the execution script Gardener can centrally control how machines are updated without the need of OS providers to (re-)implement that logic.
However, as stated in the mentioned requirements above, the execution script assumes existence of Docker and systemd.
As part of the shoot flow Gardener will create a special CRD in the seed cluster that needs to be reconciled by an extension controller, for example:
---
apiVersion: extensions.gardener.cloud/v1alpha1
kind: OperatingSystemConfig
metadata:
name: pool-01-original
namespace: default
spec:
type: <my-operating-system>
purpose: reconcile
reloadConfigFilePath: /var/lib/cloud-config-downloader/cloud-config
units:
- name: docker.service
dropIns:
- name: 10-docker-opts.conf
content: |
[Service]
Environment="DOCKER_OPTS=--log-opt max-size=60m --log-opt max-file=3"
- name: docker-monitor.service
command: start
enable: true
content: |
[Unit]
Description=Docker-monitor daemon
After=kubelet.service
[Install]
WantedBy=multi-user.target
[Service]
Restart=always
EnvironmentFile=/etc/environment
ExecStart=/opt/bin/health-monitor docker
files:
- path: /var/lib/kubelet/ca.crt
permissions: 0644
encoding: b64
content:
secretRef:
name: default-token-5dtjz
dataKey: token
- path: /etc/sysctl.d/99-k8s-general.conf
permissions: 0644
content:
inline:
data: |
# A higher vm.max_map_count is great for elasticsearch, mongo, or other mmap users
# See https://github.com/kubernetes/kops/issues/1340
vm.max_map_count = 135217728In order to support a new operating system you need to write a controller that watches all OperatingSystemConfigs with .spec.type=<my-operating-system>.
For those it shall generate a configuration blob that fits to your operating system.
For example, a CoreOS controller might generate a CoreOS cloud-config or Ignition, SLES might generate cloud-init, and others might simply generate a bash script translating the .spec.units into systemd units, and .spec.files into real files on the disk.
OperatingSystemConfigs can have two purposes which can be used (or ignored) by the extension controllers: either provision or reconcile.
- The
provisionpurpose is used by Gardener for the user-data that it later passes to the machine-controller-manager (and then to the provider's API) when creating new VMs. It contains thedownloaderunit. - The
reconcilepurpose contains the "original" user-data (that is then stored inSecrets in the shoot'skube-systemnamespace (see step 1). This is downloaded and applies late (see step 5).
As described above, the "original" user-data must be re-applicable to allow in-place updates.
The way how this is done is specific to the generated operating system config (e.g., for CoreOS cloud-init the command is /usr/bin/coreos-cloudinit --from-file=<path>, whereas SLES would run cloud-init --file <path> single -n write_files --frequency=once).
Consequently, besides the generated OS config, the extension controller must also provide a command for re-application an updated version of the user-data.
As visible in the mentioned examples the command requires a path to the user-data file.
Gardener will provide the path to the file in the OperatingSystemConfigs .spec.reloadConfigFilePath field (only if .spec.purpose=reconcile).
As soon as Gardener detects that the user data has changed it will reload the systemd daemon and restart all the units provided in the .status.units[] list (see below example). The same logic applies during the very first application of the whole configuration.
After generation extension controllers are asked to store their OS config inside a Secret (as it might contain confidential data) in the same namespace.
The secret's .data could look like this:
apiVersion: v1
kind: Secret
metadata:
name: osc-result-pool-01-original
namespace: default
ownerReferences:
- apiVersion: extensions.gardener.cloud/v1alpha1
blockOwnerDeletion: true
controller: true
kind: OperatingSystemConfig
name: pool-01-original
uid: 99c0c5ca-19b9-11e9-9ebd-d67077b40f82
data:
cloud_config: base64(generated-user-data)Finally, the secret's metadata, the OS-specific command to re-apply the configuration, and the list of systemd units that shall be considered to be restarted if an updated version of the user-data is re-applied must be provided in the OperatingSystemConfig's .status field:
...
status:
cloudConfig:
secretRef:
name: osc-result-pool-01-original
namespace: default
command: /usr/bin/coreos-cloudinit --from-file=/var/lib/cloud-config-downloader/cloud-config
lastOperation:
description: Successfully generated cloud config
lastUpdateTime: "2019-01-23T07:45:23Z"
progress: 100
state: Succeeded
type: Reconcile
observedGeneration: 5
units:
- docker-monitor.service(The .status.command field is optional and must only be provided if .spec.reloadConfigFilePath exists).
Once the .status indicates that the extension controller finished reconciling Gardener will continue with the next step of the shoot reconciliation flow.
Gardener supports specifying Container Runtime Interface (CRI) configuration in the OperatingSystemConfig resource. If the .spec.cri section exists then the name property is mandatory. The only supported values for cri.name at the moment are: containerd and docker, which uses the in-tree dockershim.
For example:
---
apiVersion: extensions.gardener.cloud/v1alpha1
kind: OperatingSystemConfig
metadata:
name: pool-01-original
namespace: default
spec:
type: <my-operating-system>
purpose: reconcile
reloadConfigFilePath: /var/lib/cloud-config-downloader/cloud-config
cri:
name: containerd
...To support ContainerD, an OS extension must :
- The operating system must have built-in ContainerD and the Client CLI
- ContainerD must listen on its default socket path:
unix:///run/containerd/containerd.sock - ContainerD must be configured to work with the default configuration file in:
/etc/containerd/config.toml(Created by Gardener).
If CRI configurations are not supported it is recommended create a validating webhook running in the garden cluster that prevents specifying the .spec.providers.workers[].cri section in the Shoot objects.
OperatingSystemConfigAPI (Golang specification)downloaderscript (fetching the "original" user-data and the execution script)- Original user-data templates
- Execution script (applying the "original" user-data)