Wave watches Deployments, StatefulSets and DaemonSets within a Kubernetes cluster and ensures that their Pods always have up to date configuration.
By monitoring mounted ConfigMaps and Secrets, Wave can trigger a Rolling Update of the Deployment when the mounted configuration is changed.
Not all software is built equal. Some applications can automatically reload their configuration and continue running with the updated configuration, while others will just continue running the old configuration until their process dies. Wave is here to help those applications that do not support dynamic configuration reloading.
When deploying applications to Kubernetes, it is common practice to deploy configuration using ConfigMaps and Secrets, mounted into the Pod either as environment variables or in files.
When this configuration is updated, if the software consuming the configuration is not able to dynamically reload, it will continue running the old configuration until the Pod is killed, this leads to inconsistency between the supposed desired state of the application and the running state of the application.
Wave monitors Deployments and their underlying configuration and will trigger the Kubernetes Deployment controller to update the application and bring up new Pods whenever the underlying configuration is changed.
This means that, with Wave, whenever a ConfigMap or Secret is updated, all Deployments mounting the configuration will bring up new Pods on the cluster and remove the old Pods with the out-of-date configuration.
Wave gives application developers confidence that the deployed configuration, matches the live configuration. It allows developers to discover misconfiguration as it is deployed, rather than when the Pods happen to be re-cycled.
Wave uses the golang Kubernetes client library which only supports the previous and the next Kubernetes version. However, since Wave only edits Deployments, Daemonsets and StatefulSet we can support older Kubernetes verions as long as no fields were removed from those three objects. You can find supported versions in the following table:
| Wave Version | API Client | Maximum Supported Kubernetes Versions | E2E Tested Versions |
|---|---|---|---|
| 0.5 | 1.14 | 1.15 | |
| 0.6+ | 1.29 | 1.30 | 1.21, 1.30 |
Wave is released periodically. The latest version is v0.8.0
A list of changes can be seen in the CHANGELOG.
Public docker images for releases since v0.4.0 are available on Quay.
Helm charts are available in this repository and hosted through Github Pages. To deploy, add the repository to helm and install:
$ helm repo add wave-k8s https://wave-k8s.github.io/wave/
$ helm install wave wave-k8s/wave
Helm will install a minimal setup. For production setups we recommend the following values:
# run two replias for HA
replicas: 2
# enable webhooks for faster updates
webhooks:
enabled: true
# make sure that replicas do not restart at the same time
pdb:
enabled: true
# schedule to multiple AZs
topologySpreadConstraints:
- maxSkew: 1
topologyKey: topology.kubernetes.io/zone
whenUnsatisfiable: DoNotSchedule
labelSelector:
matchLabels:
app: wave
# set resources. adjust this to your setup
resources:
requests:
memory: 256Mi
cpu: 25m
limits:
memory: 2Gi
Wave is a Kubebuilder based project, as such we have auto-generated Kustomize configuration as an example of how to install the controller in the config folder.
quay.io/wave-k8s/wave
If you are using RBAC within your cluster, you must grant the service account used by your Wave instance permission to read all Secrets, ConfigMaps and Deployments and the ability to update Deployments within each namespace in the cluster.
Example ClusterRole and ClusterRoleBindings are available in the
config/rbac folder or as part of the helm chart.
The following section details the various configuration options that Wave provides at the controller level.
Wave can be run in an active-standby HA configuration using Kubernetes leader election. When leader election is enabled, each Pod will attempt to become leader and, whichever is successful, will become the active or master controller. The master will perform all of the reconciliation of Resources.
The standby Pods will sit and wait until the leader goes away and then one standby will be promoted to master.
To enable leader election, set the following flags:
--leader-election=true
--leader-election-id=<name-of-leader-election-configmap>
--leader-election-namespace=<namespace-controller-runs-in>
The controller uses Kubernetes informers to cache resources and reduce load on the Kubernetes API server. Every informer has a sync period, after which it will refresh all resources within its cache. At this point, every item in the cache is queued for reconciliation by the controller.
By default, sync will happen every 10h. Kubernetes will inform Wave about changes in any Deployment, DaemonSet, StatefulSet, Secret or ConfigMap in the meantime and Wave will trigger a reconciliation right away. If you encounter any bugs you can reduce sync period by setting the following flag:
--sync-period=5m // Default value of 10h (10 hours)
This will ensure that every resource will be reconciled at least every 5 minutes. Please note that this will cause more load on your API and increase CPU load in the Wave container. It might also increase latency during the time of the sync. Do not set this unless you encounter bugs (and in that case please tell us).
You can limit Wave to only watch certain namespaces:
--namespaces=your-namespace,other-namespace
If you haven't yet got Wave running on your cluster, see Installation for details on how to get it running.
Wave watches all Deployments within a cluster but only processes those that have
the annotation wave.pusher.com/update-on-config-change: "true" which allows
individual service owners to opt-in to the Wave controller.
Therefore, to enable Wave for your Deployment, add the
wave.pusher.com/update-on-config-change annotation to your Deployment as shown
below:
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
wave.pusher.com/update-on-config-change: "true"
...
Once enabled, Wave will set the initial configuration hash as an annotation on
the Deployment's PodTemplate. You will be able to see this when reading
your Deployment:
$ kubectl get deployment foo
apiVersion: apps/v1
kind: Deployment
metadata:
name: foo
namespace: default
annotations:
wave.pusher.com/update-on-config-change: "true"
spec:
template:
metadata:
annotations:
wave.pusher.com/config-hash: "<SHA256_HASH>"
...
From now on, when a mounted ConfigMap or Secret is updated, Wave will update
this config-hash annotation and cause a Rolling Update to occur.
If your Pod is reading some ConfigMap or Secret using the API and you want it to be restarted on change you can tell Wave in an annotation:
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
wave.pusher.com/update-on-config-change: "true"
wave.pusher.com/extra-configmaps: "some-namespace/my-configmap,configmap-in-same-namespace"
wave.pusher.com/extra-secrets: "some-namespace/my-secret,some-other-namespace/foo"
...
Wave will watch those ConfigMap or Secret and behave just like if they were mounted.
This section outlines some of the underlying concepts that enable this controller to work the way it does.
Wave acts as an opt-in controller on a per Deployment basis. Before processing any Deployment, Wave checks for the presence of a "Required annotation". If the annotation is not present, Wave will ignore the Deployment.
Therefore, to enable Wave for your Deployment, add the
wave.pusher.com/update-on-config-change annotation to your Deployment as shown
below:
apiVersion: apps/v1
kind: Deployment
metadata:
annotations:
wave.pusher.com/update-on-config-change: "true"
...
Wave will now start processing this Deployment.
Wave monitors the data stored in ConfigMaps and Secrets referenced within a Deployment. By calculating a SHA256 hash of the data in a reproducible manner, Wave can determine when the data with the ConfigMaps and Secrets has changed.
Wave stores the calculated hash as an annotation on the PodTemplate within the
Deployment's specification and will update the Deployment whenever the hash is
changed.
Modifying the PodTemplate in this way causes the Kubernetes Deployment
controller to start a Rolling Update of the Deployment's Pods without changing
any of the configuration of the containers or other controllers operation on the
Pods and Deployment.
Since Wave triggers a Rolling Update you can configure how pods are replaced
in Strategy field of your Deployment object.
You can choose between RollingUpdate (default) and Recreate.
Wave watches all ConfigMaps and Secrets that are referenced by a Deployment. This allows Wave to trigger a reconciliation whenever the ConfigMaps or Secrets are modified.
Wave can update Deployments on creation/update using Mutating Webhooks.
This will prevent triggering restarts when adding the hash annotation initially.
Additionally, Wave will prevent scheduling of pods which lack any of their
required Secrets or ConfigMaps to reduce stress on the cluster.
Pods will stay in state Pending instead of ContainerCreating.
When required Secrets/ConfigMaps have been created Wave will restore the
scheduler and add the config hash without requiring any restarts.
- Found a bug? Please open an issue.
- Have a feature request. Please open an issue.
- If you want to contribute, please submit a pull request
Please see our Contributing guidelines.
This project is licensed under Apache 2.0 and a copy of the license is available here.