oneuptime
oneuptime copied to clipboard
Decrease sizing by removing useless docker pull
Is your feature request related to a problem? Please describe. As new verison of one uptime release more and more, it seems like it is starting to become heavier and heavier. I am deploying one uptime using the docker compose method on a VM. When needing to upgrade, i am unfortunately not deploying it on a new VM, checking if everything works, then destroying the old VM. I am deploying on the same VM, which means i actually need twice as many storage size in order to download all the new images. With that, i am finding quite useless to always pull postgres/clickhouse/redis image if I am already using external one. I dont know if I am the only one having this kind of issue
Describe the solution you'd like Firstly, to reduce sizing, remove useless pulling docker images when not using the container at all anyway
Describe alternatives you've considered Add some variable to enable/disable postgres/clickhouse/Redis container depending on if external variant are used instead. If you find any other container that you may thing may be useless because we could use an external variant, feel free to create a variable to disable it
Additional context Add any other context or screenshots about the feature request here.
Please run docker system prune and that should delete all the unused images.
Well the thing is, postgres/redis/clickhouse images are used, since it spawns a container, even tho functionally i am not using them. And yes, docker system prune is something i am doing using a cron that runs every 10min in case, but docker system prune doesnt work if all images are used. When deploying a new version, old images are still used until the new docker compose is running. The problem is the storage, i have to pull new verison while keeping old version