Skip to content

Latest commit

 

History

History
388 lines (311 loc) · 9.95 KB

Deployment.md

File metadata and controls

388 lines (311 loc) · 9.95 KB
Raw

Deployment Guide

1. Overview

EaseProbe is a Go application and can be deployed on any platform that supports Go.

This document describes how to deploy EaseProbe on the following ways:

  • Standalone Deployment - run EaseProbe as a systemd service.
  • Docker Deployment - run EaseProbe as a container.
  • Docker-Compose Deployment - run EaseProbe as a docker-compose service which includes a Prometheus and a Grafana.
  • Kubernetes Deployment - run EaseProbe as a kubernetes pod.
  • Helm Deployment - run EaseProbe as a kubernetes pod using helm.

2. Standalone Deployment

The following steps describe how to deploy EaseProbe as a standalone application.

2.1 Download EaseProbe

Download the latest version of EaseProbe from the EaseProbe Release Page, find the EaseProbe binary for your platform and download it.

for example,

wget https://github.com/megaease/easeprobe/releases/download/v2.0.0/easeprobe-v2.0.0-linux-amd64.tar.gz
tar -xvf easeprobe-v2.0.0-linux-amd64.tar.gz

Then, you will find a binary named easeprobe in the current ./bin directory. You can create a symbolic link to the binary.

sudo ln -sf ${PWD}/bin/easeprobe /usr/local/bin/easeprobe

2.2 Configure EaseProbe

Before running EaseProbe, you need to create a configuration file for EaseProbe. The configuration file is a YAML file.

The following is an example of a configuration file - /etc/easeprobe.conf.

http:
  - name: Google
    url: https://www.google.com

notify:
  slack:
    - name: "Slack"
      webhook: "https://hooks.slack.com/services/XXXX/BBBB/...."

settings:
  sla:
    schedule: daily
    time: 10:00:01+08:00
    data: /var/lib/easeprobe/sla #<---- data file location
  log:
    level: debug
    file: /var/log/easeprobe.log  #<---- log file location
  http:
    port: 8181
    log:
       file: /var/log/easeprobe-http-access.log #<---- access log file location

There are three parameters need your attention:

  • settings.sla.data: the data file of EaseProbe
  • settings.log.file: the log file of EaseProbe
  • settings.http.log.file: the HTTP access log file of EaseProbe

2.3 Run EaseProbe

On Linux Platform, you can configure systemd to run EaseProbe as a service.

The following is an example of a systemd service file - easeprobe.service

[Unit]
Description=EaseProbe Service
After=network.target

[Service]
Type=simple
User=root
ExecStart=/usr/local/bin/easeprobe -f /etc/easeprobe.conf
ExecStop=/bin/pkill -f easeprobe
Restart=always

[Install]
WantedBy=multi-user.target

Run the following command to install the service file.

sudo cp ./easeprobe.service /etc/systemd/system/easeprobe.service
sudo systemctl daemon-reload
sudo systemctl enable easeprobe.service
sudo systemctl start easeprobe.service

Uninstall the service file.

sudo systemctl stop easeprobe.service
sudo systemctl disable easeprobe.service
sudo rm /etc/systemd/system/easeprobe.service

3. Docker Deployment

Prepare a configuration file and a data file - config.yaml

http:
  - name: Google
    url: https://www.google.com

notify:
  slack:
    - name: "Slack"
      webhook: "https://hooks.slack.com/services/XXXX/BBBB/...."

settings:
  sla:
    schedule: daily
    time: 10:00:01+08:00

You can run the EaseProbe by the following command.

docker run -d --name easeprobe \
  -p 8181:8181 \
  -v /path/to/config.yaml:/opt/config.yaml \
  -v /path/to/data:/opt/data \
  megaease/easeprobe:latest

Note:

  • -p option is used to expose the HTTP port of EaseProbe.
  • -v option is used to mount the configuration file and data file to the container.
    • /opt/config.yaml is the configuration file default path in the container.
    • /opt/data/ is the data file default directory in the container.

4. Docker-Compose Deployment

The docker-compose deployment is a simple way to deploy EaseProbe with Prometheus and Grafana.

You can visit Docker-Compose directory to find the docker-compose file and the configuration files.

resources/docker-compose
├── compose.yaml          # <-- the docker-compose file
├── easeprobe
│   └── config.yaml       # <-- the EaseProbe configuration file
├── grafana
│   └── datasource.yaml   # <-- the Grafana datasource configuration file
└── prometheus
    └── prometheus.yaml   # <-- the Prometheus configuration file

To start the EaseProbe, Prometheus and Grafana, run the following command.

cd resources/docker-compose
docker-compose up -d

5. Kubernetes Deployment

Because EaseProbe needs to persist data, we have to deploy the EaseProbe as Stateful-Set in Kubernetes, this would lead a bit complex deployment process.

  1. Creating the ConfigMap for EaseProbe config.yaml
  2. Creating a PV/PVC for EaseProbe SLA data persistent.
  3. Deploy EaseProbe
  4. Create the EaseProbe Service

5.1 Creating the ConfigMap for EaseProbe Configuration file

The following is an example of a configuration file - config.yaml.

apiVersion: v1
kind: ConfigMap
metadata:
  name: easeprobe-conf
data:
  config.yaml: |
    http:
      - name: Google
        url: https://www.google.com
        failure: 2

      - name: Prometheus (in K8s)
        url: http://prometheus:9090/graph

      - name: ElasticSearch-01 (outside K8s)
        url: http://172.20.2.201:9200
        headers:
          Authorization: "Basic ABCDEFG1234asdf=="
    host:
      servers:
        - name: server1
          host: ubuntu@172.20.1.116
          key: /opt/login.pem

        - name: server2
          host: ubuntu@172.20.2.117
          key: /opt/login.pem
    notify:
      slack:
        - name: "MegaEase Slack#alert"
          webhook: "https://hooks.slack.com/services/ASDFA/BBBASD/....."
      discord:
        - name: "MegaEase Discord#alert"
          webhook: "https://discord.com/api/webhooks/212121212/....."
    settings:
      probe:
        interval: 1m
      log:
        level: "info"

5.2 Creating a PV/PVC for EaseProbe SLA data persistent.

To be simple, we use the NFS as an example

Step One: Create NFS PV(Persistent Volume)

Note:

You need to change the server and path to your NFS server and path.

apiVersion: v1
kind: PersistentVolume
metadata:
  name: easeprobe-storage-nfs
spec:
  capacity:
    storage: 10Gi
  volumeMode: Filesystem
  accessModes:
    - ReadWriteMany
  persistentVolumeReclaimPolicy: Recycle
  storageClassName: easeprobe-storage
  mountOptions:
    - hard
    - nfsvers=4.1
  nfs:
    path: /volumes/nfs/easeprobe
    server: 172.20.2.114

Step Two: Create NFS PVC(Persistent Volume Claim)

apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: easeprobe-pvc
spec:
  accessModes:
  - ReadWriteMany
  resources:
    requests:
      storage: 1Gi
  storageClassName: easeprobe-storage
  volumeMode: Filesystem
  volumeName: easeprobe-storage-nfs
status:
  accessModes:
  - ReadWriteMany
  capacity:
    storage: 10Gi

5.3 Deploy EaseProbe

This is the deployment file for EaseProbe.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: easeprobe
  namespace: default
spec:
  replicas: 1
  selector:
    matchLabels:
      app: easeprobe
  template:
    metadata:
      labels:
        app: easeprobe
    spec:
      containers:
      - name: easeprobe
        image: megaease/easeprobe
        ports:
        - containerPort: 8181
        volumeMounts:
          - mountPath: /opt/config.yaml
            name: configmap-volume-0
            subPath: config.yaml
          - mountPath: /opt/data
            name: pvc-volume-easeprobe-pvc
      volumes:
      - name: configmap-volume-0
        configMap:
          name: easeprobe-conf
          items:
          - key: config.yaml
            path: config.yaml
      - name: pvc-volume-easeprobe-pvc
        persistentVolumeClaim:
          claimName: easeprobe-pvc

Note:

  • The configmap-volume-0 is the ConfigMap for config.yaml, which is mounted as volume under /opt/config.yaml in the container.
  • The pvc-volume-easeprobe-pvc is the PVC for SLA data persistent, which is mounted as a volume under /opt/data in the container.

5.4 Create the EaseProbe Service

The service is used to expose the HTTP port of EaseProbe.

Note:

The following service is a ClusterIP service, you can change it to NodePort or LoadBalancer service. It uses the port 38181 to map 8181 in the container.

apiVersion: v1
kind: Service
metadata:
  name: easeprobe
  namespace: default
spec:
  ports:
  - name: httpport
    port: 38181
    protocol: TCP
    targetPort: 8181
  type: ClusterIP

6. Kubernetes Deployment Using Helm

Add repository

helm repo add easeprobe https://megaease.github.io/easeprobe

Install and run

helm install [RELEASE_NAME] easeprobe/easeprobe

Note: Persistence for EaseProbe using Helm is not enabled by default, you must enable it for production environment, refer to Helm README for more details.

Uninstall

helm uninstall [RELEASE_NAME]