What happened?
I did a fresh installation on a fresh Kubernetes cluster (1.15.5) and I get the following error in the kubernetes-networking-cluster dashboard:
Panel plugin not found: grafana-piechart-panel
It seems like the plugin is missing a simple InitContainer with the grafana-cli plugins install grafana-piechart-panel command solves this issue:
initContainers:
- image: grafana/grafana:6.4.3
name: grafana-init
command:
- bash
- -c
- grafana-cli plugins install grafana-piechart-panel
volumeMounts:
- mountPath: /var/lib/grafana
name: grafana-storage
readOnly: false
Did you expect to see some different?
I expected that Grafana has all required plugins :)
How to reproduce it (as minimally and precisely as possible):
Just use the current master and apply the changes:
ce5fe79 (HEAD -> master, origin/master, origin/HEAD) Merge pull request #299 from coreos/fix-jb
Environment
Kubernetes with kubeadm
Prometheus Operator version:
Insert image tag or Git SHA here
version: quay.io/coreos/prometheus-operator:v0.34.0
Kubernetes version information:
kubectl version
kubectl version
Client Version: version.Info{Major:"1", Minor:"15", GitVersion:"v1.15.5", GitCommit:"20c265fef0741dd71a66480e35bd69f18351daea", GitTreeState:"clean", BuildDate:"2019-10-15T19:16:51Z", GoVersion:"go1.12.10", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{Major:"1", Minor:"15", GitVersion:"v1.15.5", GitCommit:"20c265fef0741dd71a66480e35bd69f18351daea", GitTreeState:"clean", BuildDate:"2019-10-15T19:07:57Z", GoVersion:"go1.12.10", Compiler:"gc", Platform:"linux/amd64"}
Kubernetes cluster kind:
kubeadm (1.15.5)
Manifests:
insert manifests relevant to the issue
Insert Prometheus Operator logs relevant to the issue here
Insert Prometheus logs relevant to the issue here
Anything else we need to know?:
The issue is pretty simple to fix I just don't know if the patch should be provided in this repository or somewhere else.
indeed, we might need to bump the grafana version here, @pgier wdyt? :-)
Maybe let's fix dashboard coming from kubernetes-mixins to remove additional non-standard panel? This way we won't need to install panel plugin which gives us a piechart.
Additionally piechart is the most meaningless and worthless panel type and we shouldn't use it at all. More on that topic can be found in:
1) https://www.data-to-viz.com/caveat/pie.html
2) "Information Dashboard Design - Displaying data for at-a-glance monitoring" by Stephen Few, Chapter 8 - "Ideal library of graphs", Section "Dysfunctional Graphs that We Should Avoid" (p. 143).
Agreed I think the pie chart shouldn鈥檛 be there in the first place.
I just deployed master and it seems that grafana-piechart-panel is still there.
I think I just read a discussion about removing it in the upstream kubernetes-mixin project. Once that's done we can update the dependency here and we should be good.
Not that it matters if the pie charts are removed, but for future reference there's a simpler way to install Grafana plugins at runtime via the GF_INSTALL_PLUGINS environment variable:
https://grafana.com/docs/grafana/latest/installation/docker/#installing-grafana-plugins
Most helpful comment
Maybe let's fix dashboard coming from kubernetes-mixins to remove additional non-standard panel? This way we won't need to install panel plugin which gives us a piechart.
Additionally piechart is the most meaningless and worthless panel type and we shouldn't use it at all. More on that topic can be found in:
1) https://www.data-to-viz.com/caveat/pie.html
2) "Information Dashboard Design - Displaying data for at-a-glance monitoring" by Stephen Few, Chapter 8 - "Ideal library of graphs", Section "Dysfunctional Graphs that We Should Avoid" (p. 143).