ebe175afcf
fix: bump analyze timeout to 300s for 87 GB+ disk images
2026-07-21 18:26:57 +00:00
71792756ea
fix: increase analyze API timeout to 120s for large disk probing
...
guestfish needs more than 30s to probe a 7.5 GiB VMDK image
(mount filesystem, read /etc/os-release, check EFI partition).
Other endpoints keep the 30s default.
2026-07-21 18:24:26 +00:00
10199f896b
fix: escape f-string braces in backend route decorators
...
{job_id} in f-strings was interpreted as Python variable instead
of FastAPI path parameter. Double braces {{job_id}} produce
literal {job_id} in the route string.
2026-07-21 18:17:41 +00:00
af26da7b64
feat: two-phase upload/download with real-time progress + smart speed timeout
...
Split monolithic /session/start into two endpoints:
- POST /session/upload — phase 1: file acquisition only, returns JSON
- POST /session/analyze — phase 2: backend analysis, returns HTML
- GET /session/progress/{filename} — poll download progress
Uploads:
- Browser-native progress bar via XMLHttpRequest (real % + GiB)
Downloads:
- wget runs in background (Popen), frontend polls /session/progress
- HEAD request gets Content-Length before download starts
- Real-time speed (MB/s) and ETA displayed in the UI
- Smart timeout: after 30s, if ETA > 1 hour, kills download and
suggests manual download to laptop + File Upload instead
- Absolute safety net at 4 hours
UI: clear phase transitions — 'Downloading... 2.3 GiB (4.5 MB/s) ~12 min'
→ 'Step 2/2: Analysing source image...' → result
2026-07-21 18:10:52 +00:00
509eb97b57
fix: large-file support + user-defined VM name
...
Large-file handling:
- Set TMPDIR=/mnt/converter/in in service to spool uploads to
shared storage instead of 24 GB LXC rootfs (critical for >24GB)
- Chunked upload streaming (8 MiB) with progress logging every 1 GiB
- Pre-flight disk space check via Content-Length header
- Clean up partial files on upload/download failure
- Download timeout extended to 7200s (2 hours) for 88 GB images
- Switched wget from --show-progress to --progress=dot:giga
(compact output, won't fill memory on large transfers)
- uvicorn --timeout-keep-alive 300 on both frontend and backend
VM name:
- Added vm_name field to initial session form (step 1)
- Falls back to auto-generated 'os_type-vmid' if left blank
- Pre-filled & editable in confirm form (step 2)
2026-07-21 17:59:31 +00:00
4b50b185fa
feat: full deployment scripts — backend + LXC creation, frontend installer
...
backend/install.sh (run on srv2):
- Creates /mnt/converter directory tree, clones/updates git repo
- Installs 10 system packages (qemu-img, guestfish, archive tools)
- Installs Python deps + vm-bench-backend systemd service
- Interactive LXC creation: prompts for container ID, IP, gateway,
bridge, MAC (all with defaults)
- Creates vm-bench LXC: 3 cores, 6GB RAM, 24GB rootfs, nesting,
AppArmor unconfined, cgroup access, bind mount /mnt/converter
frontend/install.sh (run inside LXC):
- Installs python3, pip3, wget, curl
- Installs Python deps (fastapi, uvicorn, jinja2, etc.)
- Verifies backend connectivity, installs vm-bench systemd service
frontend/vm-bench.service: systemd unit bundled in repo
README: replaced Prerequisites/Installation with full Deployment
section covering both steps, container config table, and re-deploy
2026-07-21 17:51:53 +00:00
c887ccd5a3
fix: uniform /api/v1/ prefix, real EFI detection, spec alignment
...
Backend:
- All routes now consistently under /api/v1/ (health, analyze, jobs)
- Replace dummy EFI heuristic with actual guestfish-based detect_efi()
- Add path traversal validation on analyze source_filename
- Use typed response models (HealthResponse, CleanupResponse) everywhere
- Clean up unused imports (Path, ErrorResponse)
Models:
- Align AnalyzeResponse with open-api.yaml (add vmid, remove bootable/error,
make efi_detectable nullable, deduplicate CleanupResponse)
Converter:
- Add detect_efi(disk_path) using guestfish list-filesystems
Frontend:
- Update api_client paths to /api/v1/health and /api/v1/analyze
- Remove unsupported 'network' field from job payload
Spec:
- Promote inline HealthResponse/CleanupResponse to named schemas
- Move /analyze and /health under /api/v1/ prefix
Docs:
- Add README.md with architecture, install, API reference, usage flow
2026-07-21 17:40:08 +00:00
ab61878fbc
chore: restore open-api.yaml
2026-07-21 16:50:49 +00:00
6bcb68c552
chore: VM ID range 21000-21100, network default vmbr0, target_storage default local-lvm
2026-07-21 16:50:40 +00:00
be756e33bf
chore: add install.sh for backend systemd setup
2026-07-21 15:01:02 +00:00
439b0c51a1
chore: add systemd service file for backend deployment
2026-07-21 14:56:02 +00:00
a8099f504c
chore: initial commit — vm-bench frontend + backend
2026-07-21 14:53:29 +00:00
