KubeVela is a platform builder tool to create easy-to-use yet extensible app delivery/management systems with Kubernetes. KubeVela relies on Helm as templating engine and package format for apps. But Helm is not the only templating module that KubeVela supports. Another first-class supported approach is CUE.
Also, KubeVela is by design a Kubernetes controller (i.e. works on server side), even for its Helm part, a Helm operator will be installed.
Occasionally you might hit the issue as below. It happens when the last KubeVela release deletion hasn't completed.
Take a break and try again in a few seconds.
And manually apply all WorkloadDefinition and TraitDefinition manifests to have all capabilities back.
Occasionally you might hit the issue as below. It happens when there is an old OAM Kubernetes Runtime release, or you applied
ScopeDefinition "healthscopes.core.oam.dev" and try again.
When you look into the logs of Pod kubevela-vela-core and found the issue as below.
Error response from daemon: toomanyrequests: You have reached your pull rate limit. You may increase the limit by authenticating and upgrading: https://www.docker.com/increase-rate-limit
You can use github container registry instead.
If you hit the issue as below, an
cert-manager release might exist whose namespace and RBAC related resource conflict
Try these steps to fix the problem.
- Delete release
- Delete namespace
- Install KubeVela again
If you deploy some other services which will apply MutatingWebhookConfiguration mutating-webhook-configuration, installing KubeVela will hit the issue as below.
To fix this issue, please upgrade KubeVela Cli
vela version to be higher than
v0.2.2 from KubeVela releases.
Operating Autoscale depends on metrics server, so it has to be enabled in various clusters. Please check whether metrics server
is enabled with command
kubectl top nodes or
kubectl top pods.
If the output is similar as below, the metrics is enabled.
Or you have to manually enable metrics server in your Kubernetes cluster.
- ACK (Alibaba Cloud Container Service for Kubernetes)
Metrics server is already enabled.
- ASK (Alibaba Cloud Serverless Kubernetes)
Metrics server has to be enabled in
Operations/Add-ons section of Alibaba Cloud console as below.
Please refer to metrics server debug guide if you hit more issue.
Install metrics server as below, or you can install the latest version.
Also add the following part under
.spec.template.spec.containers in the yaml file loaded by
kubectl edit deploy -n kube-system metrics-server.
Noted: This is just a walk-around, not for production-level use.
Enable it with following command.
Have fun to set autoscale on your application.