fix(xff): truncate LEAKED IPs + ROTATION badge for rotation attacks

`for i in $(seq 1 100); do curl -H "X-Forwarded-For: 191.100.20.$i" ...`
was dumping 100 distinct IPs into AttackerDetail's LEAKED IPs row,
drowning the rest of the ORIGIN section. The 100-IP wall is itself a
signal (WAF-bypass-list probing) that deserves a short badge, not a
flood.

Backend:
- get_attacker_ip_leaks gains `limit: int = 10` parameter — caller
  only ever needs a sample, not the full set.
- New count_attacker_ip_leaks() returns the unbounded COUNT(*) via
  one cheap SQL aggregate.
- Detail endpoint returns {ip_leaks: [first 10], ip_leaks_total: N}
  so the UI can render a rotation badge independent of list length.

UI:
- New LeakedIPsRow component. First 5 distinct IPs rendered inline
  with hover tooltips (unchanged). When > 5, a `+ N more` expand
  button reveals the rest of the sample; when total exceeds the
  10-row cap, a subtle `(+M beyond sample)` note appears.
- When total ≥ 20, a red `ROTATION · N` tag renders leading the
  row with a tooltip explaining the semantic: "almost certainly
  XFF-rotation / WAF-bypass probing, not a real attribution leak."

DB churn is deliberately not capped — 100k rows × ~500 B is tolerable.
If it becomes a problem we can add an ingester-side count-and-skip;
for now the UX fix is the whole story.

Added test_ip_leaks_total_reported_separately_from_list asserting
the endpoint shape matches what the UI consumes.
This commit is contained in:
2026-04-24 18:25:46 -04:00
parent ca39552692
commit c78ab032bd
5 changed files with 204 additions and 46 deletions

View File

@@ -258,12 +258,20 @@ class BaseRepository(ABC):
raise NotImplementedError
async def get_attacker_ip_leaks(
self, attacker_uuid: str
self, attacker_uuid: str, *, limit: int = 10,
) -> list[dict[str, Any]]:
"""Return ``bounty_type='ip_leak'`` rows for the attacker, newest
first. Each row's payload carries the TCP source IP, the header
that leaked, and the claimed real IP — see the XFF-mismatch
extractor in ``decnet.web.ingester`` for the shape."""
"""Return up to ``limit`` ``bounty_type='ip_leak'`` rows for the
attacker, newest first. Each row's payload carries the TCP
source IP, the header that leaked, and the claimed real IP —
see the XFF-mismatch extractor in ``decnet.web.ingester`` for
the shape. Caller pairs with :meth:`count_attacker_ip_leaks`
to detect XFF-rotation (100+ claimed IPs from one source)."""
raise NotImplementedError
async def count_attacker_ip_leaks(self, attacker_uuid: str) -> int:
"""Total number of ``ip_leak`` bounties recorded for this
attacker. Used to detect XFF-rotation signal where the attacker
cycles through many claimed IPs (WAF-bypass-list probing)."""
raise NotImplementedError
@abstractmethod

View File

@@ -908,12 +908,14 @@ class SQLModelRepository(BaseRepository):
return [(svc, evt) for svc, evt in rows.all()]
async def get_attacker_ip_leaks(
self, attacker_uuid: str
self, attacker_uuid: str, *, limit: int = 10,
) -> list[dict[str, Any]]:
"""Return ``bounty_type='ip_leak'`` rows for this attacker, newest
first. Shape matches the XFF-mismatch payload emitted by the
ingester: keys include ``real_ip_claim``, ``source_header``,
``headers_seen``, ``path``, ``method``."""
first, capped at ``limit``. Shape matches the XFF-mismatch
payload emitted by the ingester: keys include ``real_ip_claim``,
``source_header``, ``headers_seen``. Use
:meth:`count_attacker_ip_leaks` to get the unbounded total for
rotation detection."""
async with self._session() as session:
ip_res = await session.execute(
select(Attacker.ip).where(Attacker.uuid == attacker_uuid)
@@ -926,6 +928,7 @@ class SQLModelRepository(BaseRepository):
.where(Bounty.attacker_ip == ip)
.where(Bounty.bounty_type == "ip_leak")
.order_by(desc(Bounty.timestamp))
.limit(limit)
)
out: list[dict[str, Any]] = []
for row in rows.scalars().all():
@@ -940,6 +943,22 @@ class SQLModelRepository(BaseRepository):
out.append(rec)
return out
async def count_attacker_ip_leaks(self, attacker_uuid: str) -> int:
"""Cheap COUNT(*) for XFF-rotation detection."""
async with self._session() as session:
ip_res = await session.execute(
select(Attacker.ip).where(Attacker.uuid == attacker_uuid)
)
ip = ip_res.scalar_one_or_none()
if not ip:
return 0
count_res = await session.execute(
select(func.count(Bounty.id))
.where(Bounty.attacker_ip == ip)
.where(Bounty.bounty_type == "ip_leak")
)
return int(count_res.scalar() or 0)
async def get_attacker_artifacts(self, uuid: str) -> list[dict[str, Any]]:
"""Return `file_captured` logs for the attacker identified by UUID.

View File

@@ -35,7 +35,10 @@ async def get_attacker_detail(
pairs = await repo.get_attacker_service_activity(uuid)
attacker["service_activity"] = bucket_services(pairs)
# Attribution leaks — XFF / Forwarded / X-Real-IP mismatches captured
# by the HTTP bounty extractor. Empty list when no HTTP interaction
# or no mismatch.
attacker["ip_leaks"] = await repo.get_attacker_ip_leaks(uuid)
# by the HTTP bounty extractor. Cap the returned list at 10 so a
# rotation attack (100s of forged XFF values) doesn't flood the UI;
# `ip_leaks_total` carries the unbounded count so the UI can render
# a ROTATION DETECTED badge when the count crosses a threshold.
attacker["ip_leaks"] = await repo.get_attacker_ip_leaks(uuid, limit=10)
attacker["ip_leaks_total"] = await repo.count_attacker_ip_leaks(uuid)
return attacker