[Linkerd2] 404 when access Grafana + Prometheus


#1

Hi team,

I deploy Kubernetes (1.11.2) with kubeadm on local (1 master + 1 worker node). Follow the Linkerd2 getting started link, but without create clusterrolebinding.
My K8s has enable RBAC but not deploy in GKE, so I don’t run this command.
All work fine until Step 4 (Explorer Linkerd).
I deploy via SSH because master node run on Ubuntu 16.04 Server without GUI.

I change Service type of web from ClusterIP to NodePort, then access via NodePort successfully from my PC: k8s.test.com:30157
It shows list of namespace, deployments, pods … but when I click in Grafana dashboard, the redirect link is 404 page not found in this link: k8s.test.com:30157/dashboard/db/linkerd-pod?var-namespace=linkerd&var-pod=controller-79f5959c45-ldkvd

I install K8s Dashboard, the view the log in Grafana:

....
t=2018-09-25T04:18:10+0000 lvl=info msg="Database table locked, sleeping then retrying" logger=sqlstore retry=0
t=2018-09-25T04:28:10+0000 lvl=info msg="Database table locked, sleeping then retrying" logger=sqlstore retry=0
t=2018-09-25T05:08:10+0000 lvl=info msg="Database table locked, sleeping then retrying" logger=sqlstore retry=0
t=2018-09-25T06:18:10+0000 lvl=info msg="Database table locked, sleeping then retrying" logger=sqlstore retry=0
t=2018-09-25T08:18:10+0000 lvl=info msg="Database table locked, sleeping then retrying" logger=sqlstore retry=0

Please help me how to access Grafana.
Thanks


#2

Hi @spyhunter88!

The problem has to do with the root_url that grafana is configured with. Please see this issue for details: https://github.com/linkerd/linkerd2/issues/1692