Files
DECNET/decnet/web/router/swarm/api_decommission_host.py
anti d80e6aa6d1 fix(security): close MEDIUM ASVS findings — JWT pinning, SSE tickets, SSRF, mTLS pin, rate limits + correctness bugs
Auth (V2.1.1/V3.1.2, V2.1.3, V3.1.1):
- Pin JWT iss/aud/typ at mint and require+verify them at decode; revocation
  (jti denylist + tokens_valid_from) still enforced.
- Change-password now requires min_length=12.
- SSE auth moves off JWT-in-URL to a single-use 60s opaque ticket
  (POST /auth/sse-ticket); raw JWT in query no longer authenticates a stream.
  Removed dead fail-open get_stream_user helper.

Egress (V5.1.1, V9.1.1/V14.1.3):
- Webhook delivery + CRUD reject SSRF destinations (private/loopback/link-local/
  metadata, IPv4-mapped, multi-A-record) via resolved-IP validation, pin to the
  vetted IP, and never auto-follow redirects. Opt-out via DECNET_WEBHOOK_ALLOW_PRIVATE.
- UpdaterClient pins the worker leaf cert SHA-256 against the stored per-host
  fingerprint (fail closed on missing/mismatch); DECNET_VERIFY_HOSTNAME now
  defaults True.

Hardening (V13.1.3, V4.1.4, V13.1.2):
- Rate-limit change-password (5/min), enroll-bundle (10/min), webhook-create
  (20/min), host-delete (20/min) via the existing slowapi limiter.
- Correct false 'global auth middleware' comment; document enroll-bundle proxy
  trust.

Correctness (BUG-7..11):
- BUG-7 unbound bus in finally; BUG-8 apply_ceiling clamps to min(base,ceiling);
  BUG-9 commit before emit; BUG-10 multi-actor rearm for sub-threshold identities;
  BUG-11 normalize naive timestamps to UTC.

Already-closed (no change): V14.1.1, V2.1.2/V3.1.3, V5.1.2. Tests added for
every fix; unanimous adversarial review.
2026-06-10 12:32:15 -04:00

76 lines
2.5 KiB
Python

# SPDX-License-Identifier: AGPL-3.0-or-later
"""DELETE /swarm/hosts/{uuid} — decommission a worker.
Removes the DeckyShard rows bound to the host (portable cascade — MySQL
and SQLite both honor it via the repo layer), deletes the SwarmHost row,
and best-effort-cleans the per-worker bundle directory on the master.
Also asks the worker agent to wipe its own install (keeping logs). A
dead/unreachable worker does not block master-side cleanup.
"""
from __future__ import annotations
import pathlib
from fastapi import APIRouter, Depends, HTTPException, Request, status
from decnet.logging import get_logger
from decnet.swarm.client import AgentClient
from decnet.web.db.repository import BaseRepository
from decnet.web.dependencies import get_repo, require_admin
from decnet.web.limiter import limiter
from decnet.web.router.swarm._mtls import PeerCert, require_operator_cert
log = get_logger("swarm.decommission")
router = APIRouter()
@router.delete(
"/hosts/{uuid}",
status_code=status.HTTP_204_NO_CONTENT,
tags=["Swarm Hosts"],
responses={
401: {"description": "Missing or invalid admin JWT"},
403: {"description": "Authenticated user is not an admin, or operator cert missing"},
404: {"description": "No host with this UUID is enrolled"},
429: {"description": "Too many decommission requests — retry after the window resets"},
},
)
@limiter.limit("20/minute")
async def api_decommission_host(
uuid: str,
request: Request,
repo: BaseRepository = Depends(get_repo),
_admin: dict = Depends(require_admin),
_operator: PeerCert = Depends(require_operator_cert),
) -> None:
row = await repo.get_swarm_host_by_uuid(uuid)
if row is None:
raise HTTPException(status_code=404, detail="host not found")
try:
async with AgentClient(host=row) as agent:
await agent.self_destruct()
except Exception:
log.exception(
"decommission: self-destruct dispatch failed host=%s"
"proceeding with master-side cleanup anyway",
row.get("name"),
)
await repo.delete_decky_shards_for_host(uuid)
await repo.delete_swarm_host(uuid)
# Best-effort bundle cleanup; if the dir was moved manually, don't fail.
bundle_dir = pathlib.Path(row.get("cert_bundle_path") or "")
if bundle_dir.is_dir():
for child in bundle_dir.iterdir():
try:
child.unlink()
except OSError:
pass
try:
bundle_dir.rmdir()
except OSError:
pass