Skip to content


Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time
Github PyPi Release Latest CircleCI Build Code Coverage License Python Versions Join the chat at


Pasttle is based upon, born out of the necessity to:

  • Have a lightweight pastebin software
  • Not depend on Google AE
  • Paste via CLI

Also, I wanted to play more with

Pasttle is split in server and client, it depends on what you are trying to use to run one or the other.

  • Server: Runs on python and needs a database (defaults to using SQLite) to store all the data. You want this if you are setting your own private pasttle instance
  • Client: Entirely written in bash. These are just thin wrappers around curl, you can use this to post to a pasttle server

Installing/Upgrading pasttle-server


If you are upgrading pasttle-server, you'd do well in reading CHANGELOG.rst first so you make sure the version you are upgrading has any additional steps.

Pasttle is part of PyPI, you can just:

pip install pasttle

... and it should pull all the necessary components. Whether you want to install it in a virtual environment (which I recommend) or system-wide is totally up to you.

If you instead want to try from source code (i.e. by cloning the git repo) then I suggest you execute:

python install

... and be done with it. Again, if you do this in a virtual environment or not is up to you.

In either case, if you want to add any other database drivers on top of the shipped SQLite driver, you need to install it separately depending on what method you used to install pasttle to begin with.

Running the server

Once you have it installed you need a database and (optionally) a different WSGI runner (bottle, which pasttle runs on top of, runs on wsgiref by default) but can run on Paste, tornado, CherryPy, and several others.

Also, since pasttle uses SQLAlchemy as the backend, you have your pick of any database that SQLAlchemy supports: SQLite (the default), MySQL and Postgres being the most famous.

Once you have your database all setup, you just need to create a pasttle.ini (like the one in the repo) and change the values to whatever matches your environment.

After that:

Should start the server. If you want to use a different config file, just set the environment variable PASTTLECONF to the file you want to read before starting the server, like this:

export PASTTLECONF=/etc/pasttle/mypasttle.ini
# optionally, specify a different config section, e.g. [development]
export PASTTLECONF=/etc/pasttle/mypasttle.ini:development

Alternatively, an uWSGI configuration is provided in the .ini file. Here is a script to run the server with virtualenv option:


if [ -n "$VIRTUAL_ENV" ]; then

exec uwsgi pasttle.ini --plugin python $OPT

Running via docker

There is a docker container published you can use to run pasttle, if you want to try out you can just:

docker run --rm -p 9669:9669 thekad/pasttle:latest

This pulls and runs the docker image and publishes the ports on your localhost, you can just open http://localhost:9669 at this point. If you CTRL+C your docker run it will clean everything up and leave no trace.

If you want to customize the configuration a bit, you can mount your configuration file /app/config/pasttle.ini inside the container:

docker run --rm -p 9669:9669 -v /my/config.ini:/app/config/pasttle.ini thekad/pasttle:latest

If you want to persist your data, there are a few ways to accomplish this. The first one is to run the pasttle docker image as is and make sure the sqlite database is written to a volume that can survive docker restarts, by default the docker container writes its sqlite db to /app/data so you can:

docker run --rm -p 9669:9669 -v /some/persistent/location:/app/data:rw thekad/pasttle:latest

Some people may already have a database server around, in that case you will need to consider a few things: you have to customize your config and also install the necessary driver. Here's an example running with a postgresql server using the psycopg2 driver:

bind: ; so we can publish the port outside the container
title: My dockerized pasttle
dsn: postgresql+psycopg2://
wsgi: gunicorn ; already shipped in the docker image

Then we have to run our container taking into account the build-time dependencies:

docker run --rm -p 9669:9669 -v /my/custom.ini:/app/config/pasttle.ini -e BUILD_PACKAGES="build-base postgresql-dev" -e PYTHON_PACKAGES="psycopg2" thekad/pasttle:latest

The above will install the pre-requisites to build psycopg2, then install psycopg2, and then finally run the pasttle server.

Available configuration options

dsn: <database url> [default=sqlite:///]
debug: <true/false> [default=true]
bind: <address> [default=localhost]
port: <port> [default=9669]
title: <title>
wsgi: <wsgi server to use> [default=wsgiref]
pool_recycle: <db connection age> [default=3600]
recent_items: <number to show on main page> [default=20]
pygments_style: <coloration theme> [default=tango]


See documentation of sqlalchemy.create_engine for details
WSGI server to use, look at bottle.server_names for the list
import bottle
['cgi', 'gunicorn', 'cherrypy', 'eventlet', 'tornado', 'geventSocketIO', 'rocket', 'diesel', 'twisted', 'wsgiref', 'fapws3', 'bjoern', 'gevent', 'meinheld', 'auto', 'flup', 'gae', 'paste', 'waitress']

Running the client

Running the client just requires 2 steps:

  • Source pasttle.bashrc
  • Run pasttle -h or gettle -h to check usage