I have a fundamental question. How do

dagit

and

dagster-daemon

sync with each other in case of docker based deployments? If I have

dagit

started in one container and

dagster-daemon

in the other, how does

dagit

know this is the

dagster-daemon

instance it needs to check or the other way around? What happens if I have multiple instances of

dagit

and

dagster-daemon

running in their separate containers, how does it know then which ones are for the same deployment? Does it happen solely by pointing them to the same storage/logs location?

Hi Anoop - they communicate through the database, yeah. In a dagster deployment, you can have multiple instances of dagit, but only one dagster-daemon

Okay. One of situations I can think of, where multiple dagit instances will be used, is if I have multiple repositories and want to keep them in isolated dagit environments. In such a case, if there is only one dagster-daemon instance, which repository will it pickup the schedule from?

the daemon takes in a workspace.yaml file which tells it where to find the code that it should use. Generally I think we assume that you have a single workspace.yaml that applies to all dagits and also to the daemon

But if I have multiple dagit instances, I'd probably have different location for the databases/logs for the two, right? Which one would the daemon communicate with in that case?

in that case you'd need one daemon for each dagit instance as well - the daemon needs to have a single database to talk to. It sounds like what you're describing is multiple independent deployments of dagster

That makes sense. Yeah. Just wanted to understand how the two communicated so that even if the deployments are in same machine network, it doesn't create an issue.