You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jun 1, 2024. It is now read-only.
Maintaining all these services as a big bundle of system-wide mutable state kinda sucks and we demonstrably do not have the people-power to do it. Shoving everything into containers would allow us to at least turn a system operations (which we suck at) problem largely into a software development problem (which we might be able to manage).
(I really thought there was a ticket about this open since forever, but I couldn't find it, so please feel free to dup it if I missed something obvious.)
These can all be coordinated in one big docker-compose file to run in production and we can delete just about all of the fabric stuff we currently have.
The text was updated successfully, but these errors were encountered:
Now that we're not responsible for hosting mailman, containerizing the rest of these things should be much more straightforward. For Trac / kenaan email access, we can drop the exim4 credentials into the same spot in the filesystem for talking to mailgun where they're currently present, we just have to make sure that the exim qrunner (or whatever) is running in each container so it will deliver messages enqueued with sendmail.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Maintaining all these services as a big bundle of system-wide mutable state kinda sucks and we demonstrably do not have the people-power to do it. Shoving everything into containers would allow us to at least turn a system operations (which we suck at) problem largely into a software development problem (which we might be able to manage).
(I really thought there was a ticket about this open since forever, but I couldn't find it, so please feel free to dup it if I missed something obvious.)
Sub-tasks:
These can all be coordinated in one big
docker-composefile to run in production and we can delete just about all of thefabricstuff we currently have.The text was updated successfully, but these errors were encountered: