DevOps School Bot

Manage students progress and provide the report if needed

Precondition

• PostgreSQL 13

HowTo

Start

• Add bot to the chat
• Make bot admin (to read all messages, not only commands)
• Set a webhook to bot https address: curl https://api.telegram.org/bot<BOT_TOKEN>/setWebhook\?url\=<BOT_HTTPS_URL>

via Binary

• Create a config file for bot or use ENV variables, please check the example
• Provide the db/migrations folder
• Start bot

via Docker

• docker run -it -p 3000:3000 -e DATABASE_URL="postgres_url" -e TELEGRAM_BOT_TOKEN="telegram_bot_token" to4kin/devops-school-bot:latest

via AWS Lambda

Use serverless framework to deploy AWS Lambda. Do not forget set environment variable AWSLAMBDA_ENABLED=true

service: devops-school-bot
useDotenv: true
configValidationMode: error
frameworkVersion: '>=2.61.0'

provider:
  region: "eu-central-1"
  lambdaHashingVersion: "20201221"
  name: aws
  runtime: go1.x
  logRetentionInDays: 30
  endpointType: regional
  tracing:
    apiGateway: true
    lambda: true
  ecr:
    images:
      latest:
        path: ./
  iam:
    role:
      statements:
        - Effect: "Allow"
          Resource: "*"
          Action:
            - "xray:*"

functions:
  webhook: 
    image: latest
    timeout: 15
    description: Manage students progress and provide the report if needed
    memorySize: 128
    environment:
      AWSLAMBDA_ENABLED: true
      DATABASE_URL: ${env:DATABASE_URL}
      TELEGRAM_BOT_TOKEN: ${env:TELEGRAM_BOT_TOKEN}
      
    events:
      - http:
          path: /webhook
          method: POST
          cors: false

Usage

DevOps School Bot manage students progress and provide the report

Usage:
  devops-school-bot [command]

Available Commands:
  help        Help about any command
  start       Start DevOps School Bot
  version     Print version

Flags:
  -h, --help   help for devops-school-bot

Use "devops-school-bot [command] --help" for more information about a command.

Start bot

Start DevOps School Bot with config file
Simply execute devops-school-bot start -c path/to/config/file.toml
or skip this flag to use default path

Usage:
  devops-school-bot start [flags]

Flags:
  -c, --config-path string   path to config file (default "configs/devopsschoolbot.toml")
  -h, --help                 help for start

Configuration

Availiable file formats: JSON, TOML, YAML, HCL, envfile and Java properties

example-config.toml

log_level = "info"
debug_mode = false

[apiserver]
bind_addr = ":3000"

[apiserver.cron]
enabled = true
fullreport = true
schedule = "0 15 * * FRI"

[awslambda]
enabled = false

[database]
url = "postgres://localhost/devops_school?user=postgres&password=example&sslmode=disable"
migrations = "db/migrations"

[telegram_bot]
token = "TEST_TELEGRAM_TOKEN"
verbose = false

example-config.yaml:

log_level: info
debug_mode: false

apiserver:
  bind_addr: :3000

  cron:
    enabled: true
    fullreport: true
    schedule: "0 15 * * FRI"

awslambda:
  enabled: false

database:
  url: postgres://localhost/devops_school?user=postgres&password=example&sslmode=disable
  migrations: db/migrations

telegram_bot:
  token: TEST_TELEGRAM_TOKEN
  verbose: false

Or, you can use ENV variables to update config:

LOG_LEVEL="debug"
DEBUG_MODE="false"

APISERVER_BIND_ADDR=":3000"
APISERVER_CRON_ENABLED="true"
APISERVER_CRON_FULLREPORT="true"
APISERVER_CRON_SCHEDULE="0 15 * * FRI"

AWSLAMBDA_ENABLED="false"

DATABASE_URL="postgres://localhost/devops_school?user=postgres&password=example&sslmode=disable"
DATABASE_MIGRATIONS="db/migrations"

TELEGRAM_BOT_TOKEN="TEST_TELEGRAM_TOKEN"
TELEGRAM_BOT_VERBOSE="false"

Cron job

By default, the Bot will send a full report each Friday at 15:00 UTC time to all active school chats. It's the same report as /fullreport command. If you would like to use short report (as /report command) - set fullreport = false

Docker Compose example

version: '3.7'

services:
  postgres:
    image: postgres:13.4
    container_name: postgres
    restart: always
    environment:
      POSTGRES_DB: devops_school
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: example
    volumes:
      - db_data:/var/lib/postgresql/data
  application:
    image: to4kin/devops-school-bot:latest
    container_name: devops-school-bot
    restart: always
    environment:
      DATABASE_URL: postgres://postgres/devops_school?user=postgres&password=example&sslmode=disable
      TELEGRAM_BOT_TOKEN: TELEGRAM_BOT_TOKEN
    depends_on:
      - postgres

volumes:
  db_data:

Bot commands

There're two types of telegram chat where the bot is accepting commands: private chat and group chat

User Commands

/start - Add user to database [Private chat Only]
/joinstudent - Join school as student [Group chat Only]
/joinmodule - Join school as listener [Group chat Only]
/myreport - Your progress [Private chat Only]
/homeworks - Homeworks list [Private chat Only]
/help - Help message [All chats]

Superuser Commands

/schools - Manage schools [Private chat Only]
/startschool - Start school [Group chats Only]
/stopschool - Finish school [Group chats Only]
/report - School progress [All chats]
/fullreport - School progress with homework list [All chats]

/users - Manage users [Private chat Only]
/setsuperuser - Set Superuser [Private chat Only]
/unsetsuperuser - Unset Superuser [Private chat Only]

To add homework, use the work hashtag along with the #homework, for example: #homework #cicd

For BotFather

start - Add user to database
joinstudent - Join school as student
joinmodule - Join school as listener
myreport - Your progress
homeworks - Homeworks list
help - Help message

Debug mode

When debug_mode is enabled the bot will provide additional http endpoint /debug/pprof for profiling data in the format expected by the pprof visualization tool

Student types

There're 2 types of students: full course and module course.

We have a dynamic homework list, which means that at the beginning of the school the list is empty and populated when the homework is provided by the student. This means that the student should provide all homework which was provided by other students.

On the other hand, if the student was joined for module course, the list of homework depends on the modules. The module list is also populated by students.

We need to follow one simple instruction: the homework hashtag should be #<MODULE_NAME><SEQUENTIAL_NUMBER>, for example #cicd1 #cicd2 #cloud1 #ansible1