Code for managing Cleanup of Stale Channels
You'll need to install a few libraries: pip install -r requirements.txt
You'll also need to change configuration.yaml
appropriately. warner.py
and archiver.py
should then work from the command line. Productionized automated deployments and running on a schedule are left unspecified because every production environment is unique.
That said, if you're running on Heroku, you can create a single clock
process that runs python scheduler.py
.
We recommend using virtualenv
to manage your development environment.
pip install virtualenv
virtualenv env
source env/bin/activate
In addition to requirements.txt
, there's also build-requirements.txt
for development environments. ./bin/install
will install both.
Build a local docker image:
docker build . -t destalinator
You can also use the prebuilt Docker image at randsleadershipslack/destalinator.
docker run -it -p 8080:80 destalinator sh -c "coverage html --skip-covered && python -m http.server 80"
open http://localhost:8080/htmlcov/
The Warner notifies channels that have been inactive for a period of time.
The archiver archives channels that have been inactive for a period of time.
The Announcer will notify a channel of all new channels created within a period of time.
The Flagger uses a ruleset defined in a specific channel to perform actions such as notifying channels of messages that have received a certain number of reactions.
You'll want to change this to the name of your Slack
Tune these two variables to decide how long after inactivity a channel should be warned for inactivity and then subsequently archived.
These channels need to be manually created by you in your Slack.
All configs in configuration.yaml
are overrideable through environment variables with the same name prefixed by DESTALINATOR_
(e.g. activated
-> DESTALINATOR_ACTIVATED
). Set array environment variables (e.g. DESTALINATOR_IGNORE_CHANNELS
) by comma delimiting items
- Make sure the Slackbot app is installed for your Slack
- Add a Slackbot integration, and copy the
token
parameter from the URL provided
The best way to get an API_TOKEN
is to create a new Slack App.
Once you create and name your app on your team, go to "OAuth & Permissions" to give it the following permission scopes:
channels:history
channels:read
channels:write
chat:write:bot
chat:write:user
emoji:read
users:read
After saving, you can copy the OAuth Access Token value from the top of the same screen. It probably starts with xox
.
Destalinator can be chatty and make potentially big changes to a Slack team (by warning or archiving a large amount of channels), especially when first installed.
To minimize the risk of making a mistake, Destalinator will run in a dry-run mode unless the DESTALINATOR_ACTIVATED
environment variable exists. Set it to true and Destalinator is "active." If you want to remain in dry-run mode, ensure this variable is unset/does not exist.
Tune your preferred log level for server logs or local debugging. Does not affect the ENV var specified by output_debug_env_varname
.
If you would like to log to a Slack channel as well as the default log destination, you can set true
here. The channel
will then be pulled from log_channel
in configuration.yaml
.
If you don't want to start archiving channels until a certain date right after introducing destalinator to your team,
you can set an ISO-8601 date here (YYYY-mm-dd
).
Disable the Flagger by setting DESTALINATOR_FLAGGER_DISABLED
to true
When should the destalinator run? Defaults to 4. Which time that means depends on your operating system time zone. Good luck!
Note: No DESTALINATOR_
prefix! The Sentry plugin uses this environment variable on its own.
If you would like to configure exception handling & tracking with Sentry, set up a Sentry account and configure this environment variable with the appropriate DSN value.
If you're on Heroku, you can provision this with:
heroku addons:create sentry:f1
As part of the Rands Leadership Slack community, the Rands Leadership Slack Code of Conduct applies to this project.