Report uptime as metrics for spire server and agent with config

[hixichen]

Pull Request check list

• Commit conforms to CONTRIBUTING.md?
• Proper tests/regressions included?
• Documentation updated?

Affected functionality

• Called during bootstrap.
• Add uptimeReportInterval into agent and server config.

Description of change

Add a uptime metrics for spire-agent and spire-server.
The goal here is to t indicate both Spire agent and Spire server are healthy.
And, uptime metric will help to determine if the process is periodically crashing.

This is different compared to the existing uptime in the debug API.

Test locally:

for run in {1..50}; do time make test; done

Which issue this PR fixes

fixes #1231

[amartinezfayo]

Thank you @hixichen for this contribution.
Before going into the details of the code changes, I would like to make a note that the introduction of an uptime metric has been previously discussed in #1208, where the version metric was added and adding an uptime metric was considered to be useful for the future (captured in #1231).
I just want to make sure that the motivation for this change is driven by the need of knowing the uptime specifically, and that the version metric that is emitted when the server/agent starts is not enough for your needs.
Knowing a little more about the use case and the motivation for this in your context would be helpful.
Thanks again!

[hixichen]

Thank you @hixichen for this contribution.
Before going into the details of the code changes, I would like to make a note that the introduction of an uptime metric has been previously discussed in #1208, where the version metric was added and adding an uptime metric was considered to be useful for the future (captured in #1231).
I just want to make sure that the motivation for this change is driven by the need of knowing the uptime specifically, and that the version metric that is emitted when the server/agent starts is not enough for your needs.
Knowing a little more about the use case and the motivation for this in your context would be helpful.
Thanks again!

Hi,Agustín:
Great thanks for your reply.
We are trying to use uptime to track long-term uptime since the version metric can be purged after some period of time.
Feel free to let me know if you have concerns on the code.

[hixichen]

@amartinezfayo let me know if you have other comments.

thanks.

[amartinezfayo]

Thank you @hixichen for providing the details. I think that it makes sense to add this metric to cover that case.
My main concern is around the new agent/server config settings added. Please see my comments about that.

[amartinezfayo]

I left a few nitpicks. We should be ready to go after that.
Thank you @hixichen for your patience!

[amartinezfayo]

Thanks @hixichen!

[hixichen]

Yay!
thanks. @amartinezfayo @amoore877