[Elastic Agent] Improve logging

[ruflin]

The logging output in Elastic Agent is sometimes a bit cryptic and for someone not knowing the internals of Elastic Agent, hard to interpret. Some effort should be spent going through all the log messages and evaluate how it can be improved to make it easier to understand.

Every Error message should contain details around why it is happening and potentially how it can be resolved.

[elasticmachine]

Pinging @elastic/agent (Team:Agent)

[mtojek]

One of the goal would be improving logging around enrollment. We can figure out the root cause of flakiness of elastic/elastic-package#329 .

[jlind23]

@ruflin this one seems to be to generic for me. I would rather close it and keep the following issues:

• Enhance logs when a wrong token type is used fleet-server#838
• Do not log ERROR for watchers in not upgradable scenarios  #28334
• Make enrollment errors more accurate #27764
• Fleet installation script fails to detect error in service start  elastic-agent#117

What do you think?

[elasticmachine]

Pinging @elastic/elastic-agent-control-plane (Team:Elastic-Agent-Control-Plane)

[ruflin]

SGTM. Going to close this. I also plan to add some recommendation around logging for processes run by Elastic Agent to dev docs soonish.