The installation steps requires that you have configured NFS client as described in: https://gist.github.com/jannegpriv/f78b59d93c7a815b1fbcb940277bb762
The steps below are based on information in this article.
Check for the latest official docker image for InfluxDb on DockerHub. Create an initial deployment by issuing:
k create deployment influxdb --image=docker.io/influxdb:1.7.8
The image should now be pulled from DockerHub, can take a while. You can check that the pod is running OK with:
k get pods
and check the pulling status via:
k describe pod <id of pod>
When the deployment/pod is running we will get the deployment yaml-file by issuing:
k get deployment influxdb -o yaml > influxdb-deployment.yaml
Generate secret for InfluxDb credentials:
k create secret generic influxdb-creds \
--from-literal=INFLUXDB_DATABASE=twittergraph \
--from-literal=INFLUXDB_USERNAME=root \
--from-literal=INFLUXDB_PASSWORD=root \
--from-literal=INFLUXDB_HOST=influxdb
Get and describe influxdb-creds:
# Get secret
get secret influxdb-creds
NAME TYPE DATA AGE
influxdb-creds Opaque 4 47s
# Describe secret
k describe secret influxdb-creds
Name: influxdb-creds
Namespace: default
Labels: <none>
Annotations: <none>
Type: Opaque
Data
====
INFLUXDB_DATABASE: 12 bytes
INFLUXDB_HOST: 8 bytes
INFLUXDB_PASSWORD: 4 bytes
INFLUXDB_USERNAME: 4 bytes
Trim the influxdb-deployment.yaml and add the following:
spec:
containers:
- image: docker.io/influxdb:1.7.8
imagePullPolicy: IfNotPresent
name: influxdb
envFrom:
- secretRef:
name: influxdb-creds
The influxdb-deployment.yaml should then contain:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
labels:
app: influxdb
name: influxdb
spec:
replicas: 1
selector:
matchLabels:
app: influxdb
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
labels:
app: influxdb
spec:
containers:
- image: docker.io/influxdb:1.7.8
imagePullPolicy: IfNotPresent
name: influxdb
envFrom:
- secretRef:
name: influxdb-creds
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
terminationGracePeriodSeconds: 30
Then re-create the influxdb deployment:
k delete deployment influxdb
k apply -f influxdb-deployment.yaml
# Describe the influxdb deployment, look for Environment Variables From:
k describe deployment influxdb
Name: influxdb
Namespace: default
CreationTimestamp: Mon, 30 Sep 2019 10:17:58 +0100
Labels: app=influxdb
Annotations: deployment.kubernetes.io/revision: 1
kubectl.kubernetes.io/last-applied-configuration:
{"apiVersion":"extensions/v1beta1","kind":"Deployment","metadata":{"annotations":{},"labels":{"app":"influxdb"},"name":"influxdb","namespa...
Selector: app=influxdb
Replicas: 1 desired | 1 updated | 1 total | 1 available | 0 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: app=influxdb
Containers:
influxdb:
Image: docker.io/influxdb:1.7.8
Port: <none>
Host Port: <none>
Environment Variables from:
influxdb-creds Secret Optional: false
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available True MinimumReplicasAvailable
OldReplicaSets: <none>
NewReplicaSet: influxdb-7d7c99555c (1/1 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 8m2s deployment-controller Scaled up replica set influxdb-7d7c99555c to 1
We need persistent storage for InfluxDb, create a influxdb_claim.yaml file with the following content:
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: influxdb-claim
annotations:
volume.beta.kubernetes.io/storage-class: "managed-nfs-storage"
spec:
accessModes:
- ReadWriteMany
resources:
requests:
storage: 8Gi
If you need more than 8GB storage, then change storage accordingly. Apply the PVC claim:
k apply -f influxdb-claim.yaml
Log into K3s dashboard and check for it! It should say Status: Bound if everything has gone well.
Now we must adapt the influxdb-deployment.yaml so that InfluxDb uses our PV:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
labels:
app: influxdb
name: influxdb
spec:
replicas: 1
selector:
matchLabels:
app: influxdb
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
labels:
app: influxdb
spec:
containers:
- image: docker.io/influxdb:1.7.8
imagePullPolicy: IfNotPresent
name: influxdb
volumeMounts:
- mountPath: /var/lib/influxdb
name: var-lib-influxdb
envFrom:
- secretRef:
name: influxdb-creds
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
terminationGracePeriodSeconds: 30
volumes:
- name: var-lib-influxdb
persistentVolumeClaim:
claimName: influxdb
NOTE: The claimName must match the name in the influxdb-claim.yaml file.
Re-apply the influxdb deployment to get PVs on NFS:
k apply -f influxdb-deployment.yaml
Log into K3s dashboard and check that InfluxDb deployment/pod now uses PVs.
To let Grafana "speak" to InfluxDb we need to create a Service that exposes InfluxDb pod to the cluster by issuing:
k expose deployment influxdb --port=8086 --target-port=8086 --protocol=TCP --type=ClusterIP
Get info about the svc:
# Get status of svc
k get svc influxdb
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
influxdb ClusterIP 10.43.239.12 <none> 8086/TCP 4m39s
# Describe svc
k describe svc influxdb
Name: influxdb
Namespace: default
Labels: app=influxdb
Annotations: <none>
Selector: app=influxdb
Type: ClusterIP
IP: 10.43.239.12
Port: <unset> 8086/TCP
TargetPort: 8086/TCP
Endpoints: 10.42.3.32:8086
Session Affinity: None
Events: <none>
Finally we will export the service to a file named influxdb-service.yaml:
k get svc influxdb -o yaml > influxdb-service.yaml
After trimming it looks like:
apiVersion: v1
kind: Service
metadata:
labels:
app: influxdb
name: influxdb
spec:
ports:
- port: 8086
protocol: TCP
targetPort: 8086
selector:
app: influxdb
Check for the latest official docker image for Grafana on DockerHub. Create an initial deployment by issuing:
k create deployment grafana --image=docker.io/grafana/grafana:6.3.6
The image should now be pulled from DockerHub, can take a while. You can check that the pod is running OK with:
k get pods
and check the pulling status via:
k describe pod <id of pod>
When the deployment/pod is running we will get the deployment yaml-file by issuing:
k get deployment grafana -o yaml > grafana-deployment.yaml
Generate secret for Grafana credentials:
k create secret generic grafana-creds \
--from-literal=GF_SECURITY_ADMIN_USER=admin \
--from-literal=GF_SECURITY_ADMIN_PASSWORD=grafana
Get and describe grafana-creds:
# Get secret
k get secret grafana-creds
NAME TYPE DATA AGE
grafana-creds Opaque 2 3s
# Describe secret
k describe secret grafana-creds
Name: grafana-creds
Namespace: default
Labels: <none>
Annotations: <none>
Type: Opaque
Data
====
GF_SECURITY_ADMIN_PASSWORD: 7 bytes
GF_SECURITY_ADMIN_USER: 5 bytes
Trim the grafana-deployment.yaml and add the following:
spec:
containers:
- image: docker.io/grafana/grafana:6.3.6
imagePullPolicy: IfNotPresent
name: grafana
envFrom:
- secretRef:
name: grafana-creds
The grafana-deployment.yaml should then contain:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
labels:
app: grafana
name: grafana
spec:
replicas: 1
selector:
matchLabels:
app: grafana
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
labels:
app: grafana
spec:
containers:
- image: docker.io/grafana/grafana:6.3.6
imagePullPolicy: IfNotPresent
name: grafana
envFrom:
- secretRef:
name: grafana-creds
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
terminationGracePeriodSeconds: 30
Then re-create the grafana deployment:
k delete deployment grafana
k apply -f grafana-deployment.yaml
# Describe the grafana deployment, look for Environment Variables From:
k describe deployment grafana
Name: grafana
Namespace: default
CreationTimestamp: Mon, 30 Sep 2019 13:29:11 +0100
Labels: app=grafana
Annotations: deployment.kubernetes.io/revision: 1
kubectl.kubernetes.io/last-applied-configuration:
{"apiVersion":"extensions/v1beta1","kind":"Deployment","metadata":{"annotations":{},"labels":{"app":"grafana"},"name":"grafana","namespace...
Selector: app=grafana
Replicas: 1 desired | 1 updated | 1 total | 0 available | 1 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: app=grafana
Containers:
grafana:
Image: docker.io/grafana/grafana:6.3.6
Port: <none>
Host Port: <none>
Environment Variables from:
grafana-creds Secret Optional: false
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available False MinimumReplicasUnavailable
OldReplicaSets: <none>
NewReplicaSet: grafana-5d6745f559 (1/1 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 20s deployment-controller Scaled up replica set grafana-5d6745f559 to 1
Grafana needs some configuration files, at least it needs to know how to commmunicate with InfluxDb.
That information can be stored locally in a yaml-file that via Kubernetes ConfigMap can be added to a deployment/pod.
NOTE:: The Service name influxdb is reachable via DNS if you have DNS running, can be checked via:
k get services kube-dns --namespace=kube-system
Create a influxdb-datasource.yaml file with the following content:
# config file version
apiVersion: 1
# list of datasources to insert/update depending
# what's available in the database
datasources:
# <string, required> name of the datasource. Required
- name: influxdb
# <string, required> datasource type. Required
type: influxdb
# <string, required> access mode. proxy or direct (Server or Browser in the UI). Required
access: proxy
# <int> org id. will default to orgId 1 if not specified
orgId: 1
# <string> url
url: http://influxdb:8086
# <string> database password, if used
password: root
# <string> database user, if used
user: root
# <string> database name, if used
database: influxdb
# version
version: 1
# <bool> allow users to edit datasources from the UI.
editable: true
NOTE: The url contains hostname influxdb that will point to the InfluxDb Service via DNS.
From the same directory as where the influxdb-datasource.yaml is stored, issue:
kubectl create configmap grafana-config --from-file=influxdb-datasource.yml=influxdb-datasource.yml
to create a ConfigMap.
Describe the ConfigMap grafana-config:
k describe configmap grafana-config
ame: grafana-config
Namespace: default
Labels: <none>
Annotations: <none>
Data
====
influxdb-datasource.yml:
----
# config file version
apiVersion: 1
# list of datasources to insert/update depending
# what's available in the database
datasources:
# <string, required> name of the datasource. Required
- name: influxdb
# <string, required> datasource type. Required
type: influxdb
# <string, required> access mode. proxy or direct (Server or Browser in the UI). Required
access: proxy
# <int> org id. will default to orgId 1 if not specified
orgId: 1
# <string> url
url: http://influxdb:8086
# <string> database password, if used
password: root
# <string> database user, if used
user: root
# <string> database name, if used
database: influxdb
# version
version: 1
# <bool> allow users to edit datasources from the UI.
editable: true
Events: <none>
Now we need to update the grafana.deployment.yaml file with information of the configmap.
Add the following to .spec.template.spec.volumes:
spec:
template:
spec:
volumes:
- name: grafana-config
configMap:
name: grafana-config
and the following to .spec.template.spec.containers:
spec:
template:
spec:
containers:
- name: grafana
volumeMounts:
- mountPath: /etc/grafana/provisioning/datasources/influxdb-datasource.yml
name: grafana-config
readOnly: true
subPath: influxdb-datasource.yml
The content of the grafana-deployment.yamlfile should look like this:
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
labels:
app: grafana
name: grafana
spec:
replicas: 1
selector:
matchLabels:
app: grafana
strategy:
rollingUpdate:
maxSurge: 25%
maxUnavailable: 25%
type: RollingUpdate
template:
metadata:
labels:
app: grafana
spec:
containers:
- image: docker.io/grafana/grafana:6.3.6
imagePullPolicy: IfNotPresent
name: grafana
envFrom:
- secretRef:
name: grafana-creds
volumeMounts:
- mountPath: /etc/grafana/provisioning/datasources/influxdb-datasource.yml
name: grafana-config
readOnly: true
subPath: influxdb-datasource.yml
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
dnsPolicy: ClusterFirst
restartPolicy: Always
terminationGracePeriodSeconds: 30
volumes:
- name: grafana-config
configMap:
name: grafana-config
Re-apply the grafana deployment:
k apply -f grafana-deployment.yaml
When the pod is running, check that the /etc/grafana/provisioning/datasources/influxdb-datasource.yml on the pod contains the correct info NOTE Exchange your grafana pod id in the command below (k get pods):
k exec -it grafana-b84d897fb-mj5fc cat /etc/grafana/provisioning/datasources/influxdb-datasource.yml
# config file version
apiVersion: 1
# list of datasources to insert/update depending
# what's available in the database
datasources:
# <string, required> name of the datasource. Required
- name: influxdb
# <string, required> datasource type. Required
type: influxdb
# <string, required> access mode. proxy or direct (Server or Browser in the UI). Required
access: proxy
# <int> org id. will default to orgId 1 if not specified
orgId: 1
# <string> url
url: http://influxdb:8086
# <string> database password, if used
password: root
# <string> database user, if used
user: root
# <string> database name, if used
database: influxdb
# version
version: 1
# <bool> allow users to edit datasources from the UI.
editable: true
To be able to connect to Grafana in a web browser, we need to expose a Service and an Ingress:
First create the file grafana-service.yaml with the following content:
apiVersion: v1
kind: Service
metadata:
labels:
app: grafana
name: grafana
spec:
selector:
app: grafana
ports:
- port: 3000
protocol: TCP
targetPort: 3000
name: grafana-http
and then the file grafana-traefik.yaml with the following content:
apiVersion: extensions/v1beta1
kind: Ingress
metadata:
name: openhab
annotations:
kubernetes.io/ingress.class: "traefik"
spec:
rules:
- host: k3s-grafana.example.org
http:
paths:
- path: /
backend:
serviceName: grafana
servicePort: grafana-http
Apply both files:
k apply -f grafana-service.yaml
k apply -f grafana-traefik.yaml
Then add k3s-grafana.example.org to /etc/hosts on your local computer, the surf to:
http://k3s-grafana.example.org
Log in as admin/grafana as configured earlier.
NOTE: The database is not yet created on InfluxDb, so the DS will fail to connect. Log in to the InfuxDb pod and create the database:
# Log in to pod with bash (NOTE: change to your specific pod id
k exec -it influxdb-7989845b67-29h8g /bin/bash
# Login to influxdb cmdline
root@influxdb-7989845b67-29h8g:/# influx
Connected to http://localhost:8086 version 1.7.8
InfluxDB shell version: 1.7.8
> show databases;
name: databases
name
----
_internal
>
> create database influxdb
> show databases
name: databases
name
----
_internal
influxdb
> exit