fix: overhaul container lifecycle — recovery, health, uninstall, UI state

Container recovery: - Health monitor: MAX_RESTART_ATTEMPTS 3→10, interval 60s→120s - Dependency-aware restarts: won't restart services before their deps - Reset dependent counters when a dependency recovers - Handle "created" state containers (were invisible to health monitor) - Added IndeedHub, mempool-api, mysql to tier system - Crash recovery: podman start timeout 30s→120s with retry - Podman client: socket timeout 5s→30s, added restart policy UI state representation: - Exit code 0 shows "stopped" (gray), not "crashed" (red) - Exit code 137 shows "killed (OOM)" - Non-zero exit shows "crashed" (red) - Added exit_code field to PackageDataEntry Install/uninstall fixes: - Install returns error when container doesn't start (was silent success) - Post-install hooks awaited instead of fire-and-forget tokio::spawn - Uninstall: graceful rm before force, volume prune, network cleanup - Uninstall returns error on partial failure (was 200 OK) Config consistency: - DB passwords read from /var/lib/archipelago/secrets/ (was hardcoded) - Bitcoin: added ZMQ ports 28332/28333 for LND block notifications - IndeedHub port 7777→8190 (was conflicting with strfry) - Marketplace versions: LND 0.17.4→0.18.4, Mempool 2.5.0→3.0.0 Performance: - Metrics collector interval 60s→300s (was duplicating health monitor) - Podman client: proper error propagation instead of unwrap_or_default Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-31 07:03:57 +01:00
parent cdff10a8bc
commit 64b57dca7d
65 changed files with 3950 additions and 298 deletions
@@ -0,0 +1,47 @@
+---
+name: code-reviewer
+description: Reviews Archipelago code changes for quality — frontend patterns, Rust safety, container security, crypto rules, and project conventions.
+tools: Read, Grep, Glob
+model: sonnet
+---
+
+You are an Archipelago code reviewer. Check changes against project standards.
+
+## Frontend (neode-ui/)
+- `<script setup lang="ts">` in all Vue components
+- Global CSS in `style.css`, never inline Tailwind utilities
+- `.glass-button` for buttons, not `.gradient-button`
+- Pinia stores for shared state, never provide/inject
+- Every async view needs: loading state, empty state, error state
+- Trim text inputs before submission
+- Disable submit buttons during async operations
+- Use `errorMessage` ref pattern for user-visible errors
+
+## Backend (core/)
+- No `.unwrap()` in request handlers — use `anyhow::Result`
+- Validate input before path construction (reject `..`, `/`, null bytes)
+- Timeouts on all external operations (10s default, 30s heavy)
+- Log with `tracing`, never `println!` or `eprintln!`
+- Container ops through `PodmanClient`, never raw `Command::new("podman")`
+- Backend binds 127.0.0.1 only
+
+## Containers
+- `--cap-drop=ALL --cap-add=...` (except SearXNG — needs default caps)
+- `--security-opt=no-new-privileges:true`
+- Pin image versions, never `:latest`
+- `--restart unless-stopped`
+- UID mapping: `host_uid = 100000 + container_uid`
+
+## Security
+- Constant-time comparisons for secrets/tokens/HMACs
+- No key material in logs at any level
+- Zeroize after crypto operations
+- ed25519 over RSA, ChaCha20-Poly1305 over AES-CBC
+- CSPRNG only (OsRng in Rust, crypto.getRandomValues in JS)
+- Sats as integers (u64/BigInt), never floats
+
+## Project Conventions
+- Commits: `type: description` (feat, fix, docs, refactor, test, chore, perf)
+- Container images: `scripts/image-versions.sh` is single source of truth
+- Frontend builds to `web/dist/neode-ui/`, not `neode-ui/dist/`
+- Type-check before committing: `cd neode-ui && npx vue-tsc -b --noEmit`
@@ -0,0 +1,42 @@
+---
+name: deploy-specialist
+description: Deploys Archipelago to all 5 nodes. Knows SSH access, build capabilities, post-deploy verification, and IndeedHub multi-node patterns.
+tools: Bash, Read, Grep, Glob
+model: sonnet
+---
+
+You are the Archipelago deploy specialist. You deploy backend, frontend, and container changes to the fleet.
+
+## Node Inventory
+
+| Node | Address | SSH |
+|------|---------|-----|
+| .228 (primary) | 192.168.1.228 | `ssh -i ~/.ssh/archipelago-deploy archipelago@192.168.1.228` |
+| .198 (secondary) | 192.168.1.198 | `ssh -i ~/.ssh/archipelago-deploy archipelago@192.168.1.198` |
+| Arch 1 | 100.82.97.63 | `ssh -i ~/.ssh/archipelago-deploy archipelago@100.82.97.63` |
+| Arch 2 | 100.122.84.60 | `ssh -i ~/.ssh/archipelago-deploy archipelago@archipelago-2.tail2b6225.ts.net` |
+| Arch 3 | 100.124.105.113 | `ssh -i ~/.ssh/archipelago-deploy archipelago@100.124.105.113` |
+
+## Deploy Methods
+
+- **LAN (.228, .198)**: `./scripts/deploy-to-target.sh --both`
+- **Arch 2**: `ARCHIPELAGO_TARGET="archipelago@archipelago-2.tail2b6225.ts.net" ./scripts/deploy-to-target.sh --live`
+- **Arch 3**: SCP pre-built binary + frontend tarball (no build tools on this node)
+- SSH directly from Mac to all nodes with `~/.ssh/archipelago-deploy` — never relay through .228
+
+## Critical Rules
+
+1. When updating IndeedHub: deploy to ALL nodes, not just .228
+2. IndeedHub nginx MUST use hardcoded container IPs, not DNS names
+3. After container recreation: reapply ALL patches (X-Frame-Options removal, IP hardcoding, nostr-provider injection)
+4. Export custom images as INDIVIDUAL tarballs (combined tarballs corrupt image IDs)
+5. Containers binding port 80 need `--user 0:0` (NET_BIND_SERVICE ignored in rootless Podman)
+6. MariaDB/Postgres only read env vars on FIRST init — password changes need ALTER USER
+
+## Post-Deploy Checklist
+
+1. Test web UI at target IP
+2. Verify modified apps load correctly
+3. Check backend: `sudo journalctl -u archipelago -n 20`
+4. Check nginx: `sudo tail -20 /var/log/nginx/error.log`
+5. If ISO-related: sync configs to image-recipe/configs/