Tier 1: ephemeral OpenStack staging host (boot → site.yml → smoke → destroy) #105
Etiketter
Inga etiketter
area/backups
area/ci
area/control-panel
area/identity
area/infra
area/observability
area/payments
area/security
area/storage
area/web
blocked
needs-info
needs-triage
ready-for-implementation
type
bug
type
chore
type
docs
type
epic
type
feature
type
task
wontfix
Ingen milstolpe
Inget projekt
Inga tilldelade
1 deltagare
Notiser
Förfallodatum
Inget förfallodatum satt.
Beroenden
Inga beroenden satta
Referens
bitborg/bitborg-infra#105
Läser in…
Hänvisa till i nytt ärende
Ingen beskrivning angiven.
Ta bort grenen "%!s()"
Borttagning av en gren är permanent. Även om den borttagna grenen kan fortsätta existera en kort tid innan den faktiskt tas bort, kan det INTE ångras i de flesta fall. Vill du fortsätta?
Tier 1 — on-demand prod-parity staging, no standing cost. Reuse the existing throwaway-VM machinery (ephemeral-runner ADR 0021 tofu + cloud-init; restore-drill ADR 0027 boot→do→destroy flow) to:
tofu applya single staging VM on a dedicated network/SG (staging_countvar, default 0 → $0 when idle).ansible-playbook site.ymlagainst it with thestaginginventory (from the layering task).tofu destroythe VM.Includes staging DNS + TLS: a staging domain (e.g.
*.stg.gitborg.seor an nip.io fallback) and the Let's Encrypt staging ACME directory (caddy_acme_caoverride ingroup_vars/staging) to avoid burning prod LE rate limits.Run before prod for risky changes (socket/networking/capability/app.ini). A one-shot pnpm/make wrapper + optional CI trigger.
Epic: gitborg/gitborg-docs#37
Deferred per ADR 0030 — Tier-1 full staging and the prod/staging env split are last-ditch/optional: documented as the target shape but built only when a real-host-only failure actually demands it (the restore-drill clone covers the irreversible-upgrade case first). Kept in Backlog; not scheduled. See
bitborg-docs/decisions/0030-environments-and-promotion.md.