sem-service: Managing Databases and Services on Linux#

sem-service is a utility on Linux-based virtual machines for starting, stopping, and fetching the status of background services. Started services will listen on and their default port. The IP address includes all available network interfaces. Essentially, you will be using services as if they were natively installed on the Operating System.

If you're looking to use databases in a Docker-based environment, see Working with Docker.

The general form of a sem-service command is as follows:

sem-service start [mysql | postgres | redis | memcached | mongodb | elasticsearch | rabbitmq] [version]
[--username=username] [--password=password] [--db=databasename]

Therefore, each sem-service command requires at least two parameters: the first is the task you want to perform and the second is the name of the service that will be used for the task. A third parameter is optional and is the version of the service that you want to start.
For MySQL and PostgreSQL it is possible to provide a username via --username=username, a password for the new username via --password=password, and a database name for which the user will be granted admin access via --db=dbname.

  • The default MySQL username is root, the password is blank and the default database name is test
  • The default PostgreSQL username is postgres and password is blank.

If no version value is given, a default value will be used according to the following list:

  • mysql: The default value is 5.6
  • postgres: The default value is 9.6
  • postgis: The default value is 9.6-2.5
  • redis: The default value is 4.0
  • memcached: The default value is 1.5
  • mongodb: The default value is 4.1
  • elasticsearch: The default value is 6.5
  • rabbitmq: The default is 3.8
  • cassandra: The default is 3.11
  • rethinkdb: The default is 2.3

sem-service pulls images from the Semaphore Container Registry. You can find the list of available versions on our Semaphore Container Registry images page.

The following are valid uses of sem-service:

sem-service start redis
sem-service stop redis
sem-service start redis 5
sem-service status postgres
sem-service start postgres 11
sem-service start postgres 11 --username=some_user_name --password=some_password --db=some_db_name
sem-service start postgis
sem-service start postgis 11-3.0
sem-service start mysql 8.0 --username=some_user_name --password=some_password --db=some_db_name
sem-service status mysql
sem-service start memcached
sem-service start elasticsearch
sem-service start elasticsearch 6.5
sem-service start mongodb
sem-service start mongodb 3.2
sem-service start rabbitmq
sem-service start rabbitmq 3.6
sem-service start cassandra
sem-service start cassandra 3.11.8
sem-service start rethinkdb
sem-service start rethinkdb 2.4

Services are not automatically shared across jobs in a task. To do that, start services within the prologue property of the task and populate data as needed.

Here is an example of using sem-service in a pipelines:

version: v1.0
name: Testing sem-service
    type: e1-standard-2
    os_image: ubuntu2004

  - name: Databases
        - name: DB_NAME
          value: "test"
      - name: MySQL
          - sem-service start mysql
          - sudo apt-get install -y -qq mysql-client
          - mysql --host= -uroot -e "create database $DB_NAME"
          - mysql --host= -uroot -e "show databases" | grep $DB_NAME
          - sem-service status mysql

      - name: PostgreSQL
          - sem-service start postgres
          - sudo apt-get install -y -qq postgresql-client
          - createdb -U postgres -h $DB_NAME
          - psql -h -U postgres -c "\l" | grep $DB_NAME
          - sem-service status postgres

      - name: Redis
          - sem-service start redis
          - sem-service status redis

      - name: Memcached
          - sem-service start memcached
          - sem-service status memcached