As much as “programming by recipes” can be a disaster, infrastructure recipes can be quite saver.
At the end of the day we are here because we want somebody else to be responsible for our infrastructure.
And the observability triad, perhaps, is among the top infrastructural priorities.
Now we have services like Datadog that can cover for all of the items we want in a single app.
Yet there is a fair bit of confusion of how to start using it with fly.io. I found at least 3 different topics with people reporting partial solutions.
It will be really nice to see a complete recipe of how to start forwarding logs to Datadog. I think from that point on it’ll be pretty straightforward to setup metrics and traces.