Infrastructure audit: app inventory, backups, CI, provider pricing, agent fleet #4

2026-11-04T14:46:37Z

inference-bot commented

2026-11-04 14:46:37 +00:00

Audits the operator docs against live reality (2026-11-04) and closes the gaps.

Findings the docs were wrong/stale/silent on:

No consolidated app inventory (19 apps today, was 16 at cutover); two registry paths were recorded wrongly elsewhere (portal is docs code/member-portal, panel is local/coop.inference.admin-panel).
Backups entirely undocumented (agent-host restic + Cloudron app backup; the per-user cache split and the --group-by "" retention fix).
CI (Woodpecker server + custom agent) undocumented, and currently not working: the agent is healthy but its job-collection long-poll gets HTTP 500 from the server.
Provider pricing existed only for Regolo; GreenPT/Tinfoil and the Metis decision-model route unrecorded.
No agent-fleet layout (steward + infrastructure-bot + socialbot).
Migration doc said 16 apps; the DNS split it worried about is resolved (no name served by the old box).
Token runbook said membership_sweep was nightly; it runs 3x/day.

Deliberately recorded as intentional-but-surprising: the per-user restic cache split, --group-by "", the docs/ops separation, the Tinfoil local EHBP proxy at 127.0.0.1:3301, and the pass-through metering fix.

Unresolved / needs an operator: CI server-side gRPC 500; decommission the old box; the woodpecker-agent repo has no git remote (build source only on the agent host); member-mailbox restore unknown.

Additive only; no existing sections reverted.

Audits the operator docs against live reality (2026-11-04) and closes the gaps. **Findings the docs were wrong/stale/silent on:** - No consolidated app inventory (19 apps today, was 16 at cutover); two registry paths were recorded wrongly elsewhere (portal is docs `code/member-portal`, panel is `local/coop.inference.admin-panel`). - Backups entirely undocumented (agent-host restic + Cloudron app backup; the per-user cache split and the `--group-by ""` retention fix). - CI (Woodpecker server + custom agent) undocumented, and currently **not working**: the agent is healthy but its job-collection long-poll gets HTTP 500 from the server. - Provider pricing existed only for Regolo; GreenPT/Tinfoil and the Metis decision-model route unrecorded. - No agent-fleet layout (steward + infrastructure-bot + socialbot). - Migration doc said 16 apps; the DNS split it worried about is resolved (no name served by the old box). - Token runbook said `membership_sweep` was nightly; it runs 3x/day. **Deliberately recorded as intentional-but-surprising:** the per-user restic cache split, `--group-by ""`, the docs/ops separation, the Tinfoil local EHBP proxy at 127.0.0.1:3301, and the pass-through metering fix. **Unresolved / needs an operator:** CI server-side gRPC 500; decommission the old box; the woodpecker-agent repo has no git remote (build source only on the agent host); member-mailbox restore unknown. Additive only; no existing sections reverted.

inference-bot added 1 commit 2026-11-04 14:46:37 +00:00

docs: infrastructure audit — app inventory, backups, CI, provider pricing, agent fleet 2eb88f55b4

Audit of the operator docs against live reality (2026-11-04):
- app-inventory.md: all 19 Cloudron apps with id, fqdn and registry image pin.
  Corrects the portal path (code/member-portal, not co-op/member-portal) and the
  panel path (local/coop.inference.admin-panel, not code/admin-panel).
- maintenance/backups.md: the two backup systems; the deliberate per-user restic
  cache split and the --group-by "" retention fix, with why.
- maintenance/woodpecker-ci.md: CI server + agent, image build/pin, manual-only
  trigger, git-lfs requirement, and its current not-yet-green state (agent's
  job-collection long-poll gets HTTP 500 from the server).
- provider-pricing/: index + GreenPT, Tinfoil (local EHBP proxy architecture) and
  the Metis /v1/systemone decision-model route; regolo.md updated to record the
  gateway prices were applied.
- agent-fleet.md: steward, infrastructure-bot and socialbot (Unix users, gateway
  units, teams); member-bot tier 2 planned but not yet provisioned.
- README: file list; operator-token-runbook: correct the membership_sweep cadence
  (3x daily, not nightly); migration doc: post-migration addendum (16->19 apps,
  DNS split resolved, old box still on with apps stopped).

inference-bot merged commit 6e777be842 into main

2026-11-04 14:47:18 +00:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: co-op/ops-docs#4