Hey all. Excited to be here and see if I can possibly contribute some efforts. Already doing some things locally and online in other communities, but glad to see lots of movement all around.
I have a software & design background so let me know if there is a backlog of work I can look at.
You are viewing a single comment's thread. View all
That's a tough question, mostly because I am not a Linux user outside of server environments. I guess I would have to say... _it depends_.
If you are looking to run service A, and there is some system or API dependency on service B, then you have to create some way of communicating that to the larger environment both are running in. The way I have done that in the past is via several different ways, starting with the most basic and moving into the more virtualized:
1. Write a script that will kick off service A, and then wait for that service to register itself as a PID before initiating service B script. This can be done via a simple loop that checks the system processes for a grep matching the service name, and then returning a flag to iterate into the next part of the script.
2. Virtualize each of these in a Dockerfile so that each of the scripts have to be fully kicked off before the next one starts.
3. Use a higher level language with something like Callbacks or Async code that will allow for a control flow of services.
If so, what is the best way to run a.service fully, before b.service ?
If you are looking to run service A, and there is some system or API dependency on service B, then you have to create some way of communicating that to the larger environment both are running in. The way I have done that in the past is via several different ways, starting with the most basic and moving into the more virtualized:
1. Write a script that will kick off service A, and then wait for that service to register itself as a PID before initiating service B script. This can be done via a simple loop that checks the system processes for a grep matching the service name, and then returning a flag to iterate into the next part of the script.
2. Virtualize each of these in a Dockerfile so that each of the scripts have to be fully kicked off before the next one starts.
3. Use a higher level language with something like Callbacks or Async code that will allow for a control flow of services.