minio.md

Minio

IT IS NOT RECOMMENDED TO USE MINIO IN DOCKER-COMPOSE ON PRODUCTION DEPLOYMENT. QAMETA SOFTWARE IS NOT RESPONSIBLE FOR DATA LOSS IN THIS CASE

Minio Service

minio-local:
  restart: always
  image: bitnami/minio:2022
  container_name: minio-local
  networks:
    - allure-net
  volumes:
    - minio-volume:/data
  environment:
    MINIO_ROOT_USER: ${ALLURE_S3_ACCESS_KEY}
    MINIO_ROOT_PASSWORD: ${ALLURE_S3_SECRET_KEY}
    MINIO_PROMETHEUS_AUTH_TYPE: 'public'
  ports:
    - "9000:9000"

volumes:
  minio-volume:

.env

# minio-local comes from container_name in service description above
ALLURE_S3_URL=http://minio-local:9000
ALLURE_S3_PROVIDER=s3
ALLURE_S3_BUCKET=allure-testops
ALLURE_S3_REGION=qameta-0
ALLURE_S3_ACCESS_KEY=<username>
ALLURE_S3_SECRET_KEY=<password>
ALLURE_S3_PATHSTYLE=true

Minio Provisioning Job

Do not worry when you see dead container after docker compose ps. It's job and it should die after completion. This job creates bucket in minio, so TestOps can store files in it. ENVs are used from minio service

minio-provisioning:
  restart: "no"
  image: minio/mc
  container_name: minio-provisioning
  depends_on:
    - minio-local
  networks:
    - allure-net
  entrypoint: "/bin/sh -c"
  command: >
    "mc config host add minio ${ALLURE_S3_URL} ${ALLURE_S3_ACCESS_KEY} ${ALLURE_S3_SECRET_KEY} --api S3v4 &&
    mc mb minio/${ALLURE_S3_BUCKET} --ignore-existing --region ${ALLURE_S3_REGION} &&
    mc admin info minio"

Minio Migration Job

This job is handy if you want to migrate from Local Minio to External Minio or Other S3 Storage. To perform that you need:

1. Disable LoadBalancer to stop TestOps receiving new data.
2. Wait until no jobs left in RabbitMQ (http/https)://<your_rabbit_host>:15672
3. Add service described below (minio-migrate) and add .env props
4. docker compose down
5. docker compose up -d
6. Wait until minio-migrate finishes (docker compose ps). It may take a while

minio-migrate:
  restart: "no"
  image: minio/mc
  container_name: minio-migrate
  depends_on:
    - minio
  networks:
    - allure-net
  entrypoint: "/bin/sh -c"
  command: >
    "mc config host add minio-old ${ALLURE_S3_URL} ${ALLURE_S3_ACCESS_KEY} ${ALLURE_S3_SECRET_KEY} --api S3v4 &&
     mc config host add s3-new ${S3_URL_NEW} ${ALLURE_S3_ACCESS_KEY_NEW} ${ALLURE_S3_SECRET_KEY_NEW} --api S3v4 &&
     mc cp -r minio-old/${ALLURE_S3_BUCKET}/v2 s3-new/${ALLURE_S3_BUCKET_NEW}/"

.env

S3_URL_NEW=http(s)://<new_host>:<port>
ALLURE_S3_ACCESS_KEY_NEW=<ACCESS_KEY_OF_NEW_S3>
ALLURE_S3_SECRET_KEY_NEW=<SECRET_KEY_OF_NEW_S3>
ALLURE_S3_BUCKET_NEW=<bucket_name>

FS to Minio Migration Job

This job is handy to migrate from plain file system to S3

To perform that you need:

1. Disable LoadBalancer to stop TestOps receiving new data.
2. Wait until no jobs left in RabbitMQ (http/https)://<your_rabbit_host>:15672
3. Add service described below (fs-migrate) and add .env props
4. docker compose down
5. docker compose up -d
6. Wait until fs-migrate finishes (docker compose ps). It may take a while

fs-migrate:
  restart: "no"
  image: minio/mc
  container_name: minio-fs-migrate
  networks:
    - allure-net
  entrypoint: "/bin/sh -c"
  volumes:
    ${REPORT_VOLUME}:/data
  command: >
    mc config host add s3 ${ALLURE_S3_URL} ${ALLURE_S3_ACCESS_KEY} ${ALLURE_S3_SECRET_KEY} --api S3v4 &&
    mc mb s3/${ALLURE_S3_BUCKET} --ignore-existing --region ${ALLURE_S3_REGION} &&
    mc cp -r /data/v2 s3/${ALLURE_S3_BUCKET}/

.env

ALLURE_S3_URL=http(s)://<new_host>:<port>
ALLURE_S3_ACCESS_KEY=<username>
ALLURE_S3_SECRET_KEY=<password>
ALLURE_S3_BUCKET=<bucket_name>
ALLURE_S3_REGION=<region>