Add new topic about Fleet server

[dedemorton]

Add a topic called "Fleet server" that describes:

• What is fleet-server.
• Where it lives and how it interacts with other parts of the stack? Describe architecture on a high level
• How is fleet-server run / setup
  • Prerequisites, eg. 7.13+, ECE 2.10+
  • What params exist duplicate of Document new Fleet server CLI options #527.
  • What environment variables exist - tracked in Document Fleet Server environment variables #623
  • What needs to be done in Fleet?
  • How to enroll an Elastic Agent into fleet-server - duplicate of Update existing Fleet docs for Fleet Server #522
  • How to set up multiple Fleet servers and why you would do that - tracked in Fleet scaling docs #608
  • Document Elastic Cloud and self-hosted use cases
  • ECE needs additional wildcard domains and certificates (which normally only cover *.cname, not *.*.cname) - tracked in Document Fleet Server prereqs for ECE #640
  • Document use case of missing APM server in the template for ECE Document Fleet Server prereqs for ECE #640
• Existing config options - tracked in Document Fleet Server configuration settings #642
• Compatibility: Elasticsearch >= Fleet Server >= Elastic Agent. Bugfix releases might differ. Kibana is assumed to be on the same minor as Elasticsearch. - tracked in Document stack compatibility for Fleet Server #643
• Setup in Kubernetes
• Troubleshooting: how to check logs and status endpoint /api/status

Related:

• Onboarding flow [Fleet] Define onboarding flow for fleet-server kibana#89396
• Handling TLS Handling of TLS fleet-server#90
• Renaming parameter, adding service token Add --fleet-server-service-token. Rename --fleet-server to --fleet-server-es. beats#25083
• Onboarding google doc
• Fleet server URLs [Fleet] Add fleet server URL kibana#89442

[mostlyjason]

@dedemorton Where should we document how to migrate older Elastic Agents to Fleet Server? There are a set of instructions and commands that users will need to follow. We hope to link to this from the product. elastic/kibana#95445

[mostlyjason]

Sorry I see you already answered my question here https://github.com/elastic/observability-dev/issues/1402#issuecomment-804298338. Do we need a separate tracking issue for that?

[dedemorton]

@mostlyjason To make it easier to track this work and possibly assign subsets to different writers, let's break this issue up into smaller issues.

I have an outline in google that we can use to keep track of the full list of topics.

Right now, I'm working on the first 3 items listed in this issue, which more or less map to what I intended originally when I opened this issue.

[mostlyjason]

@dedemorton sure feel free to break it up, whatever is more convenient for tracking

[dedemorton]

Basic Fleet Server docs are now merged in #585. There's still stuff to do form this list, though, so I'll keep this issue open until I either complete the items or move them to a different issue.

[dedemorton]

I'm not sure who added these items to this issue, but they need to be handled by the cloud team:

ECE needs additional wildcard domains and certificates (which normally only cover *.cname, not ..cname)
Document use case of missing APM server in the template for ECE

If you can give me more detail (link to a PR or document), I can follow up with the cloud writers.

[dedemorton]

I'm marking the following item as complete because we mention the API in a couple places in the [troubleshooting docs] (https://www.elastic.co/guide/en/fleet/master/fleet-troubleshooting.html#agent-enrollment-timeout) now:

• Troubleshooting: how to check logs and status endpoint /api/status

@mostlyjason Feel free to open an issue with details if you think the troubleshooting docs require additional info about the api.

[dedemorton]

I've either completed the work listed above or created issues to track it, so I am closing this issue now.