Meeseeker on Docker

Along with 0.0.3 release, you can now use docker to kick off meeseeker.

This will launch meeseeker in a docker container, so you can immediately attach to it on port 6380.

docker run -d -p 6380:6379 inertia/meeseeker:latest
redis-cli -p 6380

What Changed

What I did was add a Dockerfile to the project, which uses FROM redis. This means that my docker image imports everything from the redis image.

The Basic Image

After basically cloning the redis image, I have it install the packages I need, like a system version of ruby provided by apt-get:

FROM redis

ENV APP_ROOT /meeseeker
WORKDIR /meeseeker

# Dependencies
RUN \
    apt-get update && \
    apt-get install -y \
        curl \
        bzip2 \
        build-essential \
        libssl-dev \
        libreadline-dev \
        zlib1g-dev \
        ruby \
        ruby-dev

Dependencies

After we have ruby, using RUN gem update --system tells our system version of gem not to worry that I'm, running as root, which suppresses a warning when we do the next step of RUN gem install bundler, which installs the bundler utility.

Bundler is used to install the dependencies of meeseeker. Once those dependencies are installed, the docker image is ready.

RUN gem update --system
RUN gem install bundler

# copy in everything from repo
COPY . .

RUN chmod +x /meeseeker/bin/meeseeker

RUN bundle config --global silence_root_warning 1
RUN bundle install

Meanwhile, in the Container ...

The final step is to ask docker to put the image onto our docker container, which is performed with docker run above. This will actually execute the final CMD and EXPOSE statements in the Dockerfile:

CMD /usr/local/bin/redis-server --daemonize yes && bundle exec rake sync

EXPOSE 6379

Recap

So, the whole point of this Dockerfile is to get us to the final rake sync. We need all these prerequisites to happen first and docker handles this nicely.

Flexibility

You can also pass any of the environment variables meeseeker accepts. For example, this will launch meeseeker with custom_json.id channels enabled, but only keeps ops around for 5 minutes:

docker run \
  --env MEESEEKER_PUBLISH_OP_CUSTOM_ID=true \
  --env MEESEEKER_EXPIRE_KEYS=300 \
  -d -p 6380:6379 inertia/meeseeker:latest

Also see: https://hub.docker.com/r/inertia/meeseeker/

^{Previous posts:}

^{Meeseeker Update - 0.0.3pre1}
^{I'm Mr. Meeseeks! Look at me!}
^{How to do iterative steemd development on a local testnet

(shows how to set up docker for the first time)}