Verify that prereqs for running Elastic Agent and Fleet with self-managed cluster are fully documented

[dedemorton]

I'm opening this issue to alert the ingest-docs team that we are removing the docs about self-managed clusters/agents from the "ingest" topics in the Observability Guide.

Specifically, we are removing the content under the Self-managed tabs in the following topics:

• https://www.elastic.co/guide/en/observability/8.7/ingest-logs-metrics-uptime.html
• https://www.elastic.co/guide/en/observability/8.7/ingest-traces.html

We're doing this because we want to provide users with a cloud-first easy path for getting started with observability. Our rationale: Running a self-managed cluster or standalone agent is an advanced use case that is not required for users to get started with observability. It's beyond the scope of what we should cover in the obs guide.

I am opening this issue so you can verify that the Fleet guide contains all the details required by users who want to run Fleet/Elastic Agents with a self-managed cluster.

Specifically, someone needs to look at the prereqs on the self-managed tab here and confirm that any valid requirements are captured in the Fleet/Agent docs.

The self-managed instructions have not been updated in quite awhile so they should be validated with the dev or PM team.

You can see the Observability Guide updates in progress at https://github.com/elastic/observability-docs/pull/2880. We are basically renaming the section called "Send data to Elasticsearch" to "Get started with Elastic Observability" and removing all the content about Beats that basically duplicates the Beats getting started guides. Our final TOC will look something like this:

[nimarezainia]

Thanks @dedemorton the Quickstart section of the Fleet guide does have the above two topics. However they point back to the o11y for the actual content. So I guess we will have to move the content to the where rest of the Fleet/Agent dcs live. I think the content is good and those quick starts are use full.

Regarding your comment on "Running a self-managed cluster or standalone agent is an advanced use case that is not required for users to get started with observability. It's beyond the scope of what we should cover in the obs guide." - perhaps there's a disconnect, we are being asked to enhance the Standalone agent (product and docs) and make it a first class citizen for many of the o11y use cases. This may not change the plans you have above but it is something that we (on the ingest side) are prioritizing. I suspect that o11y docs will need to refer back to ingest at some point we are done fxing Standalone.

(fyi @vinaychandrasekhar @ruflin )

[ruflin]

We need to separate two topics here:

• Self managed Elasticsearch cluster
• Standalone Elastic Agent

We are on the same page around self manage Elasticsearch cluster, this is an advanced used case. Standalone Elastic Agent isn't and shouldn't be. Starting with standalone Elastic Agent should be as simple as it is today with Filebeat: Pick a log path, point it to Elasticsearch, done. This is even simpler then having to go through Fleet, onboard agents etc.

[dedemorton]

I probably should not have even mentioned standalone agent in this issue because it's really about removing the steps about self-managed clusters that really don't belong in the observability guide. AFAIK the onboarding experience for Observability is still focused around using integrations and Fleet with ESS, so that's what we want to cover in the observability docs. We want users to see how easy it is to get started with ESS in cloud!

we are being asked to enhance the Standalone agent (product and docs) and make it a first class citizen for many of the o11y use cases

That's good to hear! I've been advocating for that approach for quite awhile. I hope we can put some resources behind the effort and make the experience great for users. The docs need a lot of love.