Runbook
All website and tooling is Megan’s. Justin does no site admin or changes. For rare problems Megan cannot handle, the fallback is emailing Daryle (the prior site contractor; he also holds the Divi lifetime license). This is used very infrequently.
Site is down or broken
Section titled “Site is down or broken”- Confirm scope: whole site down, or one page/section? Check on a phone (cellular, not the office network) and a second browser to rule out local issues.
- Check the HostArmada status page and the Cloudflare status page.
- If a recent paste caused it: revert by pasting the previous known-good HTML back into the Divi Code Module (see Deploy process), or restore from UpdraftPlus.
- If the host is up but the site errors: clear the LiteSpeed cache from WordPress admin.
Hosting and the domain registrar are HostArmada (both have logins).
Cloudflare holds DNS and is shared access under Justin’s account. Account
pointers are kept off this site in
notes/current/reference/accounts-access-internal.md.
Rent tool is failing
Section titled “Rent tool is failing”- A
RATE_LIMITEDresponse is expected behavior, not an outage (RentCast monthly cap). The page should still show the official HUD figure. - Real failure: redeploy from
workers/rent-lookup/withnpx wrangler deploy, check secrets are set. See Rent Lookup worker.
Forms not sending
Section titled “Forms not sending”WPForms submissions are delivered to both Justin’s and Megan’s email and synced to Mailchimp. If leads stop arriving, check all three. Form IDs and which page uses which form are in Forms & links; confirm the form still exists in WPForms admin and the shortcode/ID on the page matches.
A leaked secret
Section titled “A leaked secret”The only secrets in the codebase are the rent worker’s HUD_API_TOKEN and
RENTCAST_API_KEY, and they live in Cloudflare, not git. Rotate per
Rent Lookup worker. For any other leaked
credential, rotate at the provider.
Backups and disaster recovery
Section titled “Backups and disaster recovery”- WordPress: UpdraftPlus runs daily and weekly. Megan also takes manual backups around larger changes.
- Large changes go through a staging site before production.
- Recovery order if rebuilding from scratch: registrar/host (HostArmada) → DNS (Cloudflare) → WordPress restore from UpdraftPlus → re-paste content from the repo → redeploy the rent worker.