Client Requests / REQ-8841
REQ-8841 · Restore a site from backup
Organisation Fernwood Creative
Site / environment lumenpress.example — Production (cp-40021)
Reported down since 20 Sep 2026, 9:15 AM
Reported change Plugin or theme update
Customer note "Updated the storefront theme this morning, then the homepage started showing a white screen."
Customer's suggested copy Their manual backup from 18 Sep 2026, 7:04 PM
Created 20 Sep 2026, 9:22 AM by customer
cp-40021 — Backups mirrors the real Environment → Backups panel
Taken Kind Status
18 Sep 2026, 7:04 PM
Manual (customer)
Available Restore…
2 Sep 2026, 3:10 AM
Scheduled (platform)
Retention expired Restore…
If the timing doesn't line up with either copy, "Import site archive" accepts a ZIP the customer supplies — from a plugin export or local copy — and stages it as a restore point.
What's real vs. proposed here
Verified: Client Requests queue Ops has a real "Client Requests" section listing Reference / Type / Title / Organisation / Status / Created, open-and-in-progress by default.
Verified: Backups panel & actions Per-environment Backups panel, "Create backup," "Import site archive," and per-backup "Restore…" / "Delete…" are real operator actions we saw live. We also saw a real backup flagged "passed its retention period, so the copy it held has been removed from storage" — modelled here as the expired row.
Verified: async job model Ops runs backup/restore as tracked background jobs (job types include backup_create and backup_restore, with Pending/Running/Complete/Failed/Rolled-back states and an actor) — not shown as its own panel here, but it's the real mechanism a "Restore…" click would kick off.
Proposed: this request detail page We didn't find a per-request detail view in ops, only the list. Everything that turns REQ-8841 into a page — the structured incident summary, embedding the Backups panel inline, and "Resolve & notify" — is proposed.
Proposed: request ↔ job ↔ customer-status link The biggest gap: nothing we observed ties a client request, a backup/restore job, and the customer-visible timeline (step 3) together. Today these look like three separate systems (requests list, backups panel, jobs queue). Wiring them together is the main build item this journey depends on — everything else is presentation.