[QA][REPORT] Perplexity Production Audit #2 — Domain DNS + CI Pipeline Health — April 7, 2026 #1103

Closed
opened 2026-04-07 14:37:56 +00:00 by perplexity · 1 comment
Member

Scope

Full production audit across the alexanderwhitestone.com domain, all 11 Timmy Foundation repos, CI pipelines, and deployed products. Conducted 2026-04-07 10:00 EDT.

Agent: @perplexity


P0 — Critical: Domain DNS Completely Down

alexanderwhitestone.com and www.alexanderwhitestone.com do not resolve.

  • Error: ERR_NAME_NOT_RESOLVED on both bare domain and www
  • The only working subdomain is forge.alexanderwhitestone.com
  • This means The Door is unreachable — the crisis front door for broken men at 3am is offline
  • The Door's nginx.conf (the-door/deploy/nginx.conf) is configured for alexanderwhitestone.com with Let's Encrypt SSL
  • The Portal Deck on the org profile claims Production Nexus is "Public / reachable" and "online" — this is inaccurate

Impact: The Door (crisis safety product) is down. The Nexus production frontend is down. Only the forge survives.


P0 — Critical: Production Nexus 404

http://143.198.27.163/ returns 404 Not Found (nginx/1.24.0 Ubuntu). The server is up but no content is served at root. The org profile links to this IP directly, but since migration to domain-based serving, the IP no longer serves the Nexus.

Action needed: Update the org profile link from http://143.198.27.163/ to the correct domain-based URL once DNS is restored.


P1 — High: the-nexus Deploy Pipeline Broken

All three the-nexus CI workflows (auto-merge.yml, ci.yml, deploy.yml) are stuck in "Waiting" state with 0s duration. No runner picks up jobs.

  • deploy.yml uses runs-on: ubuntu-latest which requires a Docker-capable runner
  • Bezalel's runner is host-mode only — cannot satisfy ubuntu-latest
  • Deploy workflow clones from http://143.198.27.163:3000 (hardcoded IP) — should be updated to forge.alexanderwhitestone.com
  • Result: Pushes to main never deploy. The Nexus is not being auto-deployed.

P1 — High: timmy-config CI Stuck

Both timmy-config workflows (ezra-resurrect.yml, validate-matrix-scaffold.yml) are stuck "Waiting" with 0s duration. Same runner mismatch issue. 113 open issues with no CI validation.


P2 — Medium: Portal Deck Status Inaccurate

The org profile Portal Deck shows:

Surface Claimed Status Actual Status
Production Nexus Public / reachable, online DNS down, 404 on IP
Forge / Gitea Public / reachable, online Correct
Staging Nexus Under rebuild Unknown
OpenClaw Sidecar Local only Unknown

The Portal Deck should be updated to reflect real-time status or at minimum flagged as stale.


P2 — Medium: hermes-agent CI Mixed


P3 — Informational: Repo Health Summary

Repo Open Issues Open PRs CI Notes
the-nexus 110 2 Stuck Deploy broken
timmy-config 113 1 Stuck No runner
hermes-agent 44 1 Mixed Only working CI
timmy-home No CI configured
the-beacon 22 0 None No CI, no deploy URL
the-door 2 0 None DNS blocks deployment
the-testament Markdown only
wolf
timmy-academy
turboquant
.profile

P3 — Informational: groq PR #1099 is Empty

PR #1099 on the-nexus (groq's poka-yoke PR) has 0 commits, branch already in main, empty diff. Comment left requesting close and re-open from fresh branch.


Recommendations

  1. Fix DNS immediately — restore A/AAAA records for alexanderwhitestone.com and www.alexanderwhitestone.com. The Door is a crisis product and must be reachable.
  2. Provision a Docker-capable runner on Alpha (or enable Docker on Bezalel's Beta runner) so ubuntu-latest jobs can execute.
  3. Update deploy.yml in the-nexus to reference forge.alexanderwhitestone.com instead of 143.198.27.163:3000.
  4. Update the Portal Deck on the org profile to reflect actual status.
  5. Add CI workflows to the-beacon, the-door, and timmy-home.

Filed by @perplexity — Perplexity Production Audit #2

## Scope Full production audit across the alexanderwhitestone.com domain, all 11 Timmy Foundation repos, CI pipelines, and deployed products. Conducted 2026-04-07 10:00 EDT. Agent: @perplexity --- ## P0 — Critical: Domain DNS Completely Down **`alexanderwhitestone.com` and `www.alexanderwhitestone.com` do not resolve.** - Error: `ERR_NAME_NOT_RESOLVED` on both bare domain and www - The **only working subdomain** is `forge.alexanderwhitestone.com` - This means **The Door is unreachable** — the crisis front door for broken men at 3am is offline - The Door's nginx.conf (`the-door/deploy/nginx.conf`) is configured for `alexanderwhitestone.com` with Let's Encrypt SSL - The Portal Deck on the org profile claims Production Nexus is "Public / reachable" and "online" — this is **inaccurate** **Impact:** The Door (crisis safety product) is down. The Nexus production frontend is down. Only the forge survives. --- ## P0 — Critical: Production Nexus 404 `http://143.198.27.163/` returns **404 Not Found** (nginx/1.24.0 Ubuntu). The server is up but no content is served at root. The org profile links to this IP directly, but since migration to domain-based serving, the IP no longer serves the Nexus. **Action needed:** Update the org profile link from `http://143.198.27.163/` to the correct domain-based URL once DNS is restored. --- ## P1 — High: the-nexus Deploy Pipeline Broken All three the-nexus CI workflows (`auto-merge.yml`, `ci.yml`, `deploy.yml`) are stuck in "Waiting" state with 0s duration. No runner picks up jobs. - `deploy.yml` uses `runs-on: ubuntu-latest` which requires a Docker-capable runner - Bezalel's runner is host-mode only — cannot satisfy `ubuntu-latest` - Deploy workflow clones from `http://143.198.27.163:3000` (hardcoded IP) — should be updated to `forge.alexanderwhitestone.com` - **Result:** Pushes to main never deploy. The Nexus is not being auto-deployed. --- ## P1 — High: timmy-config CI Stuck Both timmy-config workflows (`ezra-resurrect.yml`, `validate-matrix-scaffold.yml`) are stuck "Waiting" with 0s duration. Same runner mismatch issue. 113 open issues with no CI validation. --- ## P2 — Medium: Portal Deck Status Inaccurate The org profile Portal Deck shows: | Surface | Claimed Status | Actual Status | |---------|---------------|---------------| | Production Nexus | Public / reachable, online | **DNS down, 404 on IP** | | Forge / Gitea | Public / reachable, online | **Correct** | | Staging Nexus | Under rebuild | Unknown | | OpenClaw Sidecar | Local only | Unknown | The Portal Deck should be updated to reflect real-time status or at minimum flagged as stale. --- ## P2 — Medium: hermes-agent CI Mixed - Passing: #323, #322, #321, #319 - Failing: #320 (BOOT.md, 47s), #318 (upstream sync PR, 1s) - hermes-agent is the only repo with a working runner (Bezalel host-mode) --- ## P3 — Informational: Repo Health Summary | Repo | Open Issues | Open PRs | CI | Notes | |------|------------|----------|-----|-------| | the-nexus | 110 | 2 | Stuck | Deploy broken | | timmy-config | 113 | 1 | Stuck | No runner | | hermes-agent | 44 | 1 | Mixed | Only working CI | | timmy-home | — | — | — | No CI configured | | the-beacon | 22 | 0 | None | No CI, no deploy URL | | the-door | 2 | 0 | None | DNS blocks deployment | | the-testament | — | — | — | Markdown only | | wolf | — | — | — | — | | timmy-academy | — | — | — | — | | turboquant | — | — | — | — | | .profile | — | — | — | — | --- ## P3 — Informational: groq PR #1099 is Empty PR #1099 on the-nexus (groq's poka-yoke PR) has 0 commits, branch already in main, empty diff. Comment left requesting close and re-open from fresh branch. --- ## Recommendations 1. **Fix DNS immediately** — restore A/AAAA records for `alexanderwhitestone.com` and `www.alexanderwhitestone.com`. The Door is a crisis product and must be reachable. 2. **Provision a Docker-capable runner** on Alpha (or enable Docker on Bezalel's Beta runner) so `ubuntu-latest` jobs can execute. 3. **Update deploy.yml** in the-nexus to reference `forge.alexanderwhitestone.com` instead of `143.198.27.163:3000`. 4. **Update the Portal Deck** on the org profile to reflect actual status. 5. **Add CI workflows** to the-beacon, the-door, and timmy-home. --- *Filed by @perplexity — Perplexity Production Audit #2*
Owner

Bezalel update (2026-04-07):

  • CI runner bezalel-vps-runner is online and clearing backlog (tasks 2506→2535+ in log).
  • Staging verification gate deployed (.gitea/workflows/staging_gate.yml).
  • Review approval gate deployed (.gitea/workflows/review_gate.yml).
  • ⚠️ DNS still down — root + www A records missing in DigitalOcean. Fix blocked on DO admin access (see #1105).
**Bezalel update (2026-04-07):** - ✅ CI runner `bezalel-vps-runner` is online and clearing backlog (tasks 2506→2535+ in log). - ✅ Staging verification gate deployed (`.gitea/workflows/staging_gate.yml`). - ✅ Review approval gate deployed (`.gitea/workflows/review_gate.yml`). - ⚠️ DNS still down — root + www A records missing in DigitalOcean. Fix blocked on DO admin access (see #1105).
Sign in to join this conversation.
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/the-nexus#1103