Allow modules to run only on specific worker

[dklimpel]

Fixes: #10701

This is not the requested solution of #10701
I am not sure if a module should decide independently where it runs.

This makes it possible to configure the workers at a central location (homerserver.yaml).
I am not sure if a config value is needed to run a module on all servers with the same configuration.

Pull Request Checklist

• Pull request is based on the develop branch
• Pull request includes a changelog file. The entry should:
  • Be a short description of your change which makes sense to users. "Fixed a bug that prevented receiving messages from other servers." instead of "Moved X method from EventStore to EventWorkerStore.".
  • Use markdown where necessary, mostly for code blocks.
  • End with either a period (.) or an exclamation mark (!).
  • Start with a capital letter.
  • Feel free to credit yourself, by adding a sentence "Contributed by @github_username." or "Contributed by [Your Name]." to the end of the entry.
• Pull request includes a sign off
• Code style is correct
  (run the linters)

Signed-off-by: Dirk Klimpel [email protected]

[babolivier]

I am not sure if a module should decide independently where it runs.

This is kind of already what we do with ModuleApi.looping_background_call, where we allow the module to decide whether to run its background process on all of the processes or just the one in charge of background jobs.

Personally I'm more keen to delegate the logic of figuring out what to run on which worker. The module interface rewrite was in part motivated by the fact that the old module system would force you to write split your module if you wanted part of it to implement interface X and part of it to implement interface Y (which meant the only way to implement both at once was to get really hacky, e.g. sharing state via global variables, etc).
It feels like going backwards to me to force module developers to split their module just because they don't want to run method A and method B on the same worker. I think the modules themselves should have the possibility to decide which worker each functionality runs on, just like they have the possibility to decide which callback categories they want to implement.

[anoadragon453]

Additionally, this solution as it stands only allows for running a module on a single worker, whereas there's often cases where modules should run across a set of multiple workers.

[dklimpel]

Additionally, this solution as it stands only allows for running a module on a single worker, whereas there's often cases where modules should run across a set of multiple workers.

IMO you can run with this solution a module on multiple workers:
e.g.:

modules:
  - module: my_super_module.MySuperClass
    worker_name: Worker1
  - module: my_super_module.MySuperClass
    worker_name: Worker2

There were a chance to improve this solution:

• set worker_name to None run on all workers incl. main
• set worker_name to main run only on main

[babolivier]

(sorry, I missclicked)

[dklimpel]

I close the PR and wait for another solution. If this kind of solution has a chance, I can reopen it.