Look into ways of improving logs for users running a node

[hstove]

This is a placeholder issue, more work needs to be done to figure out exactly what work should be done.

For many people who run a node for the first time, there are a lot of logs, and often those logs make it seem like something is broken even when everything is behaving normally. In some cases this is due to a warn used when a debug might be better, or info for something that should maybe be a debug. There are also instances where a specific line is logged much too frequently.

Improving logs could make it easier to run a node and better debug when things go wrong.

[AshtonStephens]

Is this ticket consensus release critical for Nakamoto?

[hstove]

@AshtonStephens nope, it's just to improve signer onboarding

[hstove]

One small change we could make it to support different log levels. Right now, you can only use either trace, debug, or fatal log levels specified via ENV vars. We could also support warn and error by changing this code:

stacks-core/stacks-common/src/util/log.rs

Lines 247 to 259 in ae5b716

	fn inner_get_loglevel() -> slog::Level {
	if env::var("STACKS_LOG_TRACE") == Ok("1".into()) {
	slog::Level::Trace
	} else if env::var("STACKS_LOG_DEBUG") == Ok("1".into())
	|| env::var("BLOCKSTACK_DEBUG") == Ok("1".into())
	{
	slog::Level::Debug
	} else if env::var("STACKS_LOG_CRITONLY") == Ok("1".into()) {
	slog::Level::Critical
	} else {
	slog::Level::Info
	}
	}

[AshtonStephens]

Yeah actually that would be a great fast improvement, I see we have warn! a lot in the code.