Hermes VM
|
2fc23d6baa
|
feat(vm): fix devops-backend VM module — Phase 0.1 complete
- Switch backend runner from node:20-alpine to node:20-slim so GNU df
flags (--output=pcent/avail) work inside the container
- Add volume mounts to docker-compose.yml: scripts (ro), VM logs (rw),
docker.sock; set VM_SCRIPTS_PATH + VM_LOG_DIR env vars
- Rebuild repository.ts: env-configurable paths, cron history parser,
unhealthy-container inspector, Ollama model endpoints
- Add routes: GET /api/vm/cron-status, unhealthy containers, Ollama
models, container restart, model unload
- vm-cleanup.sh: add step_cosmos_pglog, step_docker_aged_images; fix
(( count++ )) → count=$(( count + 1 )) for set -e compatibility
- Add docs/VM_OBSERVABILITY_ROADMAP.md
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
2026-05-27 21:13:45 +00:00 |
|
Hermes VM
|
d0b8ce2c74
|
feat: add VM Health page to devops dashboard
Backend (Fastify):
- New module: modules/vm/ (types, repository, routes)
- GET /api/vm/health — runs vm-health-check.sh --json, returns structured result
- GET /api/vm/cleanup-log — tails /var/log/vm-cleanup.log
- POST /api/vm/cleanup — triggers vm-cleanup.sh (weekly / monthly / dry-run)
- Registered vmRoutes in server.ts
Frontend (Next.js):
- New page: /vm — VM Health
- Overall status banner (OK/WARN/CRIT) with issue summary
- Per-check cards: disk, load, RAM, swap, crash loops, container health,
build cache, docker images, journal, syslog — color-coded by level
- Cleanup trigger buttons (dry-run, weekly, monthly) with output viewer
- Collapsible cleanup log viewer (last 40 lines)
- Auto-refresh every 60s
- sidebar-nav.tsx: added 'VM Health' entry with Server icon
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
2026-05-27 18:53:20 +00:00 |
|