Skip to content

docs(onprem): make the self-hosting page a real landing page - #15464

Merged
devGregA merged 1 commit into
DefectDojo:bugfixfrom
devGregA:devgrega/docs-onprem-landing
Aug 1, 2026
Merged

docs(onprem): make the self-hosting page a real landing page#15464
devGregA merged 1 commit into
DefectDojo:bugfixfrom
devGregA:devgrega/docs-onprem-landing

Conversation

@devGregA

Copy link
Copy Markdown
Contributor

installation_options.md was two sentences and a sales address. That was fine when it was one of three pages in the section; it isn't now that it sits above pages covering sizing, installation, migration, upgrades, backups, upload limits, and storage.

It now does what a landing page should:

  • Explains the two deployment models and what separates them. A single Compose host has to be sized for peak rather than average, because nothing in that arrangement scales out and the peak is usually a large import landing while people are working in the UI. Kubernetes runs the same components as separate workloads, so you provision for steady state and add replicas when load arrives. The page gives a recommendation instead of leaving the reader to work it out, and it says a single host is a legitimate choice rather than a compromise.
  • States the database position: external managed for production, which is what the Helm chart assumes, with the containerized option called out as suitable for evaluation and not for production data.
  • Covers what you need before starting, including that sizing depends on both finding count and concurrent users because those drive different parts of the deployment, and how to obtain a license and the tooling.
  • Keeps both contact paths. The original page sent everyone to the sales address. Existing customers now get their account representative or support, and evaluators still get sales. Sending a prospect to a support address would have been a regression.

Notes for review

Frontmatter is deliberately untouched. The diff hunks start at line 8, so this does not collide with the weight: 4 to weight: 5 change in #15459. The two PRs can merge in either order.

No cross-links yet. Most of the pages this one describes are still in open PRs, so linking them would fail the offline link check on this base. The section navigation covers discovery in the meantime, and the links belong in a single pass once #15452, #15455, #15456, #15458, #15459, and #15461 have landed.

Testing

  • hugo --minify --gc --config config/production/hugo.toml builds clean with no new warnings.
  • No dead internal links. Both email links render through the theme's obfuscation hook with noscript fallbacks, verified in the built output.

🤖 Generated with Claude Code

The page was two sentences and a sales address, which no longer works now
that it sits above a section covering sizing, installation, migration,
upgrades, backups, and storage.

It now does the job a landing page should. It explains the two deployment
models and what actually separates them, that a single Compose host has to
be sized for peak because nothing scales out while Kubernetes lets you
provision for steady state and add replicas, and it gives a recommendation
rather than leaving the reader to infer one.

Adds what you need before starting, including the point that sizing depends
on both finding count and concurrent users because those drive different
parts of the deployment, and how to obtain a license and tooling.

Keeps both contact paths. Existing customers are pointed at their account
representative or support, and evaluators at the sales address the page
previously used on its own.

Frontmatter is deliberately untouched so this does not collide with the
weight change in the hardware sizing PR.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@devGregA
devGregA merged commit 143e622 into DefectDojo:bugfix Aug 1, 2026
286 of 405 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants