blumeops/pulumi/gandi/__main__.py

105 lines
3 KiB
Python
Raw Normal View History

"""Pulumi program to manage eblu.me DNS via Gandi LiveDNS.
This program manages DNS records for blumeops infrastructure:
- Wildcard record for *.ops.eblu.me pointing to indri's Tailscale IP
- indri hosts Caddy as the reverse proxy for all services
- This allows services to be accessed via real DNS names while remaining
tailnet-only (Tailscale IPs are not publicly routable)
Authentication:
Set GANDI_PERSONAL_ACCESS_TOKEN environment variable.
See docs/how-to/configuration/rotate-gandi-pat.md for PAT management.
"""
import os
import socket
import pulumi
import pulumiverse_gandi as gandi
# Get configuration
config = pulumi.Config()
domain = config.require("domain") # eblu.me
subdomain = config.require("subdomain") # ops
# Resolve indri's Tailscale IP dynamically via MagicDNS
# This script runs on the tailnet, so we can resolve the hostname directly.
# indri hosts Caddy, which reverse-proxies all services.
# Break-glass: set BLUMEOPS_REVERSE_PROXY_IP env var to override DNS resolution
REVERSE_PROXY_HOST = "indri.tail8d86e.ts.net"
tailscale_ip = os.environ.get("BLUMEOPS_REVERSE_PROXY_IP") or socket.gethostbyname(
REVERSE_PROXY_HOST
)
# Wildcard A record for *.ops.eblu.me
# Points to indri's Tailscale IP, which is only routable within the tailnet.
# This allows containers and other systems to resolve real DNS names
# while keeping services private to the tailnet.
wildcard_record = gandi.livedns.Record(
"ops-wildcard",
zone=domain,
name=f"*.{subdomain}",
type="A",
ttl=300,
values=[tailscale_ip],
)
# Base subdomain record (ops.eblu.me) - same IP
base_record = gandi.livedns.Record(
"ops-base",
zone=domain,
name=subdomain,
type="A",
ttl=300,
values=[tailscale_ip],
)
# ============== Public Services (Fly.io proxy) ==============
# CNAME records pointing public subdomains to Fly.io for reverse proxying
# back to the tailnet. See docs/how-to/expose-service-publicly.md
docs_public = gandi.livedns.Record(
"docs-public",
zone=domain,
name="docs",
type="CNAME",
ttl=300,
values=["blumeops-proxy.fly.dev."],
)
cv_public = gandi.livedns.Record(
"cv-public",
zone=domain,
name="cv",
type="CNAME",
ttl=300,
values=["blumeops-proxy.fly.dev."],
)
Expose Forgejo publicly at forge.eblu.me (#278) ## Summary Expose Forgejo publicly at `forge.eblu.me` via the Fly.io reverse proxy — the first dynamic, authenticated public-facing service. - **Forgejo hardening:** Domain changed to forge.eblu.me, SSH stays on forge.ops.eblu.me, reverse proxy trust headers configured, local registration locked to external-only (Authentik SSO) - **Tailscale Ingress:** ExternalName Service + Ingress in tailscale-operator creates forge.tail8d86e.ts.net endpoint - **Fly.io proxy:** nginx server block with rate-limited auth endpoints (3r/s), fail2ban with custom nginx-deny action, security headers, /swagger blocked, WebSocket support, 512m body limit - **Authentik:** OAuth callback updated to forge.eblu.me - **DNS/TLS:** CNAME record in Pulumi, cert in fly-setup - **Rename:** ~29 files updated from forge.ops.eblu.me to forge.eblu.me (HTTPS refs only; SSH, container builds, and Caddy table kept as-is) ## Deployment Order 1. `mise run provision-indri -- --tags forgejo` (config changes) 2. Verify forge.ops.eblu.me still works 3. `argocd app set tailscale-operator --revision feature/forge-public && argocd app sync tailscale-operator` 4. Verify `curl https://forge.tail8d86e.ts.net` 5. `cd fly && fly deploy` 6. Verify pre-DNS: `curl -H "Host: forge.eblu.me" https://blumeops-proxy.fly.dev/` 7. `fly certs add forge.eblu.me -a blumeops-proxy` 8. `argocd app set authentik --revision feature/forge-public && argocd app sync authentik` 9. `mise run dns-preview && mise run dns-up` 10. Full verification (see below) 11. Rehearse `mise run fly-shutoff` 12. After merge: reset ArgoCD revisions to main, re-sync ## Verification Checklist - [ ] forge.eblu.me loads, shows public repos - [ ] forge.ops.eblu.me still works from tailnet - [ ] SSH clone via forge.ops.eblu.me:2222 works - [ ] HTTPS clone via forge.eblu.me works - [ ] UI shows forge.eblu.me for HTTPS clone, forge.ops.eblu.me for SSH - [ ] /swagger returns 403 - [ ] Rapid login attempts trigger 429 rate limit - [ ] fail2ban bans after 5 failed logins in 10 minutes - [ ] ArgoCD can still sync (SSH unaffected) - [ ] `mise run fly-shutoff` stops all public traffic - [ ] `mise run services-check` passes Reviewed-on: https://forge.eblu.me/eblume/blumeops/pulls/278
2026-03-03 08:40:41 -08:00
forge_public = gandi.livedns.Record(
"forge-public",
zone=domain,
name="forge",
type="CNAME",
ttl=300,
values=["blumeops-proxy.fly.dev."],
)
C1: deploy adelaide-baby-shower-app to ringtail k3s (#349) ## Summary Brings up the Adelaide / Heidi / Addie baby shower app on ringtail k3s with the public/private split that the app's hosting contract calls for: `shower.eblu.me` (public, via Fly proxy) and `shower.ops.eblu.me` (tailnet). App is consumed as a wheel from the Forgejo PyPI index — source lives at [`adelaide-baby-shower-app`](https://forge.eblu.me/eblume/adelaide-baby-shower-app). ### What's included - **ArgoCD app + manifests** under `argocd/manifests/shower/` (deployment, service, ProxyGroup ingress, ConfigMap for `DJANGO_DEBUG`/`DJANGO_ADMIN_URL`, ExternalSecret for `DJANGO_SECRET_KEY` from 1Password item `Shower (blumeops)`, NFS PV on sifaka, RWX media PVC, RWO local-path data PVC for SQLite). Recreate rollout because SQLite is single-writer. - **Public surface** (`fly/`): new `shower.eblu.me` server block proxying to `shower.ops.eblu.me`. `/admin/` returns 403 at the edge except `/admin/login/` and `/admin/logout/`, which are rate-limited via a new `shower_auth` zone. `X-Clacks-Overhead` on. GNU Terry Pratchett. - **fail2ban** filter (`shower-admin-login.conf`) matching 401/403/429 on `/admin/login/` and jail (`shower.conf`) with `maxretry=5/findtime=600/bantime=3600`. The `nginx-deny` action was generalized to take a per-jail `nginx_deny_file` so the shower has its own deny list (forge keeps using the legacy default). - **Caddy** route on indri (`shower.ops.eblu.me` → `https://shower.tail8d86e.ts.net`). - **Pulumi** Gandi CNAME `shower.eblu.me → blumeops-proxy.fly.dev.`. - **Grafana** APM dashboard `configmap-shower-apm.yaml` (request rate, error rate, failed admin login count, latency percentiles, bandwidth, access logs) mirroring `docs-apm.json` with a `host="shower.eblu.me"` filter. - **Container** `containers/shower/default.nix` — `dockerTools.buildLayeredImage` with a nixpkgs Python and a startup wrapper that creates `/app/data/.venv`, pip-installs `adelaide-baby-shower-app==1.0.0` from the forge PyPI index on first boot, runs migrations + collectstatic, and execs gunicorn. A `local_settings.py` shim pins `DATABASES.NAME`/`MEDIA_ROOT`/`STATIC_ROOT` to absolute paths so they don't end up in site-packages. - **Docs** runbook at `docs/how-to/operations/shower-app.md` linked from the apps registry, plus changelog fragments. ### Defense layers on the public surface 1. fly nginx geo+fail2ban `$shower_banned` (per-service deny list) 2. fly nginx `limit_req zone=shower_auth` (3 r/s per Fly-Client-IP) 3. django-axes (5 fails / 1h, keyed on username+ip_address) 4. edge `/admin/` block (returns 403 for anything that isn't login/logout) ## Prerequisites for the user to do (NOT in this PR) Halted on these per request — they touch shared/manual systems: - [x] **NFS share** on sifaka: `/volume1/shower`, NFS rule for ringtail RW, `chown 1000:1000` - [ ] **1Password item** `Shower (blumeops)` in the blumeops vault with a freshly minted `secret-key` field (`openssl rand -base64 48`) — do NOT reuse anything that has lived in git - [ ] **Container build**: `mise run container-build-and-release shower`, then update `images[].newTag` in `argocd/manifests/shower/kustomization.yaml` to the resulting `v1.0.0-<sha>-nix` - [x] **DNS**: `mise run dns-up` after merge - [x] **Fly cert**: `fly certs add shower.eblu.me -a blumeops-proxy` - [ ] **Caddy push**: `mise run provision-indri -- --tags caddy` - [ ] **Fly redeploy** to pick up the new nginx block + fail2ban jail: `mise run fly-deploy` - [ ] **ArgoCD sync**: `argocd app set shower --revision shower-app-deploy && argocd app sync shower` to test from this branch before merging ## Test plan - [ ] Container builds successfully on nix-container-builder runner - [ ] Pod starts, migrations run, gunicorn answers on :8000 - [ ] `kubectl --context=k3s-ringtail -n shower logs deploy/shower` clean - [ ] `curl -sf https://shower.ops.eblu.me/` returns the splash page (tailnet) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/` returns 200 (pre-DNS verification) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/users/` returns 403 (edge block) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/login/` returns a Django login response - [ ] After DNS is up: `curl -I https://shower.eblu.me/` returns 200 with `X-Clacks-Overhead` - [ ] Grafana dashboard "Shower APM" appears and starts showing traffic - [ ] `mise run services-check` passes Reviewed-on: https://forge.eblu.me/eblume/blumeops/pulls/349
2026-05-11 13:47:18 -07:00
shower_public = gandi.livedns.Record(
"shower-public",
zone=domain,
name="shower",
type="CNAME",
ttl=300,
values=["blumeops-proxy.fly.dev."],
)
# ============== Exports ==============
pulumi.export("domain", domain)
pulumi.export("wildcard_fqdn", f"*.{subdomain}.{domain}")
pulumi.export("base_fqdn", f"{subdomain}.{domain}")
pulumi.export("target_ip", tailscale_ip)
pulumi.export("docs_public_fqdn", f"docs.{domain}")
pulumi.export("cv_public_fqdn", f"cv.{domain}")
Expose Forgejo publicly at forge.eblu.me (#278) ## Summary Expose Forgejo publicly at `forge.eblu.me` via the Fly.io reverse proxy — the first dynamic, authenticated public-facing service. - **Forgejo hardening:** Domain changed to forge.eblu.me, SSH stays on forge.ops.eblu.me, reverse proxy trust headers configured, local registration locked to external-only (Authentik SSO) - **Tailscale Ingress:** ExternalName Service + Ingress in tailscale-operator creates forge.tail8d86e.ts.net endpoint - **Fly.io proxy:** nginx server block with rate-limited auth endpoints (3r/s), fail2ban with custom nginx-deny action, security headers, /swagger blocked, WebSocket support, 512m body limit - **Authentik:** OAuth callback updated to forge.eblu.me - **DNS/TLS:** CNAME record in Pulumi, cert in fly-setup - **Rename:** ~29 files updated from forge.ops.eblu.me to forge.eblu.me (HTTPS refs only; SSH, container builds, and Caddy table kept as-is) ## Deployment Order 1. `mise run provision-indri -- --tags forgejo` (config changes) 2. Verify forge.ops.eblu.me still works 3. `argocd app set tailscale-operator --revision feature/forge-public && argocd app sync tailscale-operator` 4. Verify `curl https://forge.tail8d86e.ts.net` 5. `cd fly && fly deploy` 6. Verify pre-DNS: `curl -H "Host: forge.eblu.me" https://blumeops-proxy.fly.dev/` 7. `fly certs add forge.eblu.me -a blumeops-proxy` 8. `argocd app set authentik --revision feature/forge-public && argocd app sync authentik` 9. `mise run dns-preview && mise run dns-up` 10. Full verification (see below) 11. Rehearse `mise run fly-shutoff` 12. After merge: reset ArgoCD revisions to main, re-sync ## Verification Checklist - [ ] forge.eblu.me loads, shows public repos - [ ] forge.ops.eblu.me still works from tailnet - [ ] SSH clone via forge.ops.eblu.me:2222 works - [ ] HTTPS clone via forge.eblu.me works - [ ] UI shows forge.eblu.me for HTTPS clone, forge.ops.eblu.me for SSH - [ ] /swagger returns 403 - [ ] Rapid login attempts trigger 429 rate limit - [ ] fail2ban bans after 5 failed logins in 10 minutes - [ ] ArgoCD can still sync (SSH unaffected) - [ ] `mise run fly-shutoff` stops all public traffic - [ ] `mise run services-check` passes Reviewed-on: https://forge.eblu.me/eblume/blumeops/pulls/278
2026-03-03 08:40:41 -08:00
pulumi.export("forge_public_fqdn", f"forge.{domain}")
C1: deploy adelaide-baby-shower-app to ringtail k3s (#349) ## Summary Brings up the Adelaide / Heidi / Addie baby shower app on ringtail k3s with the public/private split that the app's hosting contract calls for: `shower.eblu.me` (public, via Fly proxy) and `shower.ops.eblu.me` (tailnet). App is consumed as a wheel from the Forgejo PyPI index — source lives at [`adelaide-baby-shower-app`](https://forge.eblu.me/eblume/adelaide-baby-shower-app). ### What's included - **ArgoCD app + manifests** under `argocd/manifests/shower/` (deployment, service, ProxyGroup ingress, ConfigMap for `DJANGO_DEBUG`/`DJANGO_ADMIN_URL`, ExternalSecret for `DJANGO_SECRET_KEY` from 1Password item `Shower (blumeops)`, NFS PV on sifaka, RWX media PVC, RWO local-path data PVC for SQLite). Recreate rollout because SQLite is single-writer. - **Public surface** (`fly/`): new `shower.eblu.me` server block proxying to `shower.ops.eblu.me`. `/admin/` returns 403 at the edge except `/admin/login/` and `/admin/logout/`, which are rate-limited via a new `shower_auth` zone. `X-Clacks-Overhead` on. GNU Terry Pratchett. - **fail2ban** filter (`shower-admin-login.conf`) matching 401/403/429 on `/admin/login/` and jail (`shower.conf`) with `maxretry=5/findtime=600/bantime=3600`. The `nginx-deny` action was generalized to take a per-jail `nginx_deny_file` so the shower has its own deny list (forge keeps using the legacy default). - **Caddy** route on indri (`shower.ops.eblu.me` → `https://shower.tail8d86e.ts.net`). - **Pulumi** Gandi CNAME `shower.eblu.me → blumeops-proxy.fly.dev.`. - **Grafana** APM dashboard `configmap-shower-apm.yaml` (request rate, error rate, failed admin login count, latency percentiles, bandwidth, access logs) mirroring `docs-apm.json` with a `host="shower.eblu.me"` filter. - **Container** `containers/shower/default.nix` — `dockerTools.buildLayeredImage` with a nixpkgs Python and a startup wrapper that creates `/app/data/.venv`, pip-installs `adelaide-baby-shower-app==1.0.0` from the forge PyPI index on first boot, runs migrations + collectstatic, and execs gunicorn. A `local_settings.py` shim pins `DATABASES.NAME`/`MEDIA_ROOT`/`STATIC_ROOT` to absolute paths so they don't end up in site-packages. - **Docs** runbook at `docs/how-to/operations/shower-app.md` linked from the apps registry, plus changelog fragments. ### Defense layers on the public surface 1. fly nginx geo+fail2ban `$shower_banned` (per-service deny list) 2. fly nginx `limit_req zone=shower_auth` (3 r/s per Fly-Client-IP) 3. django-axes (5 fails / 1h, keyed on username+ip_address) 4. edge `/admin/` block (returns 403 for anything that isn't login/logout) ## Prerequisites for the user to do (NOT in this PR) Halted on these per request — they touch shared/manual systems: - [x] **NFS share** on sifaka: `/volume1/shower`, NFS rule for ringtail RW, `chown 1000:1000` - [ ] **1Password item** `Shower (blumeops)` in the blumeops vault with a freshly minted `secret-key` field (`openssl rand -base64 48`) — do NOT reuse anything that has lived in git - [ ] **Container build**: `mise run container-build-and-release shower`, then update `images[].newTag` in `argocd/manifests/shower/kustomization.yaml` to the resulting `v1.0.0-<sha>-nix` - [x] **DNS**: `mise run dns-up` after merge - [x] **Fly cert**: `fly certs add shower.eblu.me -a blumeops-proxy` - [ ] **Caddy push**: `mise run provision-indri -- --tags caddy` - [ ] **Fly redeploy** to pick up the new nginx block + fail2ban jail: `mise run fly-deploy` - [ ] **ArgoCD sync**: `argocd app set shower --revision shower-app-deploy && argocd app sync shower` to test from this branch before merging ## Test plan - [ ] Container builds successfully on nix-container-builder runner - [ ] Pod starts, migrations run, gunicorn answers on :8000 - [ ] `kubectl --context=k3s-ringtail -n shower logs deploy/shower` clean - [ ] `curl -sf https://shower.ops.eblu.me/` returns the splash page (tailnet) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/` returns 200 (pre-DNS verification) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/users/` returns 403 (edge block) - [ ] `curl -I -H "Host: shower.eblu.me" https://blumeops-proxy.fly.dev/admin/login/` returns a Django login response - [ ] After DNS is up: `curl -I https://shower.eblu.me/` returns 200 with `X-Clacks-Overhead` - [ ] Grafana dashboard "Shower APM" appears and starts showing traffic - [ ] `mise run services-check` passes Reviewed-on: https://forge.eblu.me/eblume/blumeops/pulls/349
2026-05-11 13:47:18 -07:00
pulumi.export("shower_public_fqdn", f"shower.{domain}")