Hello, is there any way to avoid downtime when restarting the platform? Any idea or strategy that can guide on how to achieve it?
Any plans on adding logic to support maintenance HTMLs from nginx container?
It really very much depends how much changes you are considering, and there is no general solution. For instance, if your upgrade includes non backward-compatible database migrations, then there must be some downtime.
According to this StackOverflow question it should be possible to restart a container with no downtime with the --build option.
Can you please check whether this solution works? If it does, please open a github issue and we’ll implement it, as no-downtime deployment is an important feature.
Any plans on adding logic to support maintenance HTMLs from nginx container?
This is not on our roadmap, but it should not be too hard to implement as part of a tutor plugin.
1 Like