DS3Lab / DBCourseMaterial

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Data Modeling and Databases (DMDB)

This repository hosts public course material for the Data Modeling and Databases course at ETH given by Prof. Ce Zhang. The main purpose is to host the interactive tutorials on Katacoda.

Alternatives to Katacoda

Docker

Prerequisites:

Start a Postgres server:

docker run -d \
    -p 5432:5432 \
    -v /:/host/:ro \
    --name postgres-server \
    -e POSTGRES_HOST_AUTH_METHOD=trust \
    postgres:13.2-alpine

Nothing else should be required. This launches a docker container in the background. You can see the running containers with the following command:

docker ps

The container you just run should show up in the output (called postgres-server). You can now stop and start that container with the following commands:

docker stop postgres-server
docker start postgres-server

This should preserve the databases you have created inside the container. If you want to remove it completely (including the databases), use the following:

docker rm postgres-server

You can run the standard Postgres command line client inside that container like this (setting the user to postgres):

docker exec -it postgres-server psql -U postgres

The same way, you can start a shell inside that container:

docker exec -it postgres-server psql -U bash

To use pgcli instead, run the following:

docker run --rm -it \
    --link postgres-server \
    -v /:/host/:ro \
    -v $HOME/.config/pgcli/:/root/.config/pgcli/ \
    ds3lab/pgcli -h postgres-server -U postgres

Both containers are run such that the entire host file system is available (read-only) under /host/ inside of the containers. If you need to import files inside the containers, access the files accordingly. Also, for pgcli, the config folder of your home directory is mounted into the container such that the configuration and command history is shared with the host.

pgcli via pip

Prerequisites:

You may also be able to install pgcli on your own machine. Normally, the following is enough:

pip3 install pgcli

However, you need a basic compiler infrastructure and the Postgres client library installed. On Debian, Ubuntu, etc, the following may do that:

sudo apt install build-essential libpq-dev

Alternatively, you may be lucky with installing an old version:

pip install pgcli==2.1.1 --only-binary psycopg2

About


Languages

Language:Shell 90.5%Language:Dockerfile 9.5%