merge: testing → main (reconcile 2-week divergence)

This commit is contained in:
2026-04-28 18:36:00 -04:00
parent 499836c9e4
commit 862e4dbb31
1235 changed files with 160255 additions and 7996 deletions

View File

@@ -0,0 +1,23 @@
ARG BASE_IMAGE=debian:bookworm-slim
FROM ${BASE_IMAGE}
RUN apt-get update && apt-get install -y --no-install-recommends \
python3 \
&& rm -rf /var/lib/apt/lists/*
COPY syslog_bridge.py /opt/syslog_bridge.py
COPY server.py /opt/server.py
COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh
EXPOSE 110 995
RUN useradd -r -s /bin/false -d /opt logrelay \
&& apt-get update && apt-get install -y --no-install-recommends libcap2-bin \
&& rm -rf /var/lib/apt/lists/* \
&& (find /usr/bin/ -maxdepth 1 -name 'python3*' -type f -exec setcap 'cap_net_bind_service+eip' {} \; 2>/dev/null || true)
HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
CMD kill -0 1 || exit 1
USER logrelay
ENTRYPOINT ["/entrypoint.sh"]

View File

@@ -0,0 +1,3 @@
#!/bin/bash
set -e
exec python3 /opt/server.py

View File

@@ -0,0 +1,120 @@
#!/usr/bin/env python3
"""
Per-instance stealth seeding for honeypot service templates.
The whole decoy fleet looks identical to a scanner unless each decky
diverges on the boring details: cluster UUIDs, auth salts, uptime, minor
version strings, etc. This module derives a stable per-instance seed
from NODE_NAME (+ optional INSTANCE_ID) and exposes helpers that return
deterministic-per-decky-but-different-across-the-fleet values.
Connection-time jitter is intentionally NOT seeded — two hits to the same
decky should not replay the same latency curve.
"""
from __future__ import annotations
import asyncio
import hashlib
import os
import random
import time
import uuid
from typing import Sequence, TypeVar
T = TypeVar("T")
_HOSTNAME = (
os.environ.get("NODE_NAME")
or os.environ.get("HOSTNAME")
or "decky"
)
_INSTANCE_ID = os.environ.get("INSTANCE_ID", "")
_SEED_MATERIAL = f"{_HOSTNAME}:{_INSTANCE_ID}".encode()
_SEED_INT = int.from_bytes(hashlib.sha256(_SEED_MATERIAL).digest()[:8], "big")
#: Deterministic RNG seeded per decky — use for *persistent* choices
#: (versions, UUIDs, stored credentials). Never use for timing.
rng = random.Random(_SEED_INT)
#: Process boot time — real uptime elapsed since container start.
_PROCESS_START = time.time()
#: Deterministic per-instance fake "has been up for this long at boot"
#: offset, so every decky pretends to have a different history.
_BOOT_OFFSET = rng.randint(3600, 45 * 86400)
def hostname() -> str:
return _HOSTNAME
def uptime_seconds() -> int:
"""Monotonically increasing, unique per instance."""
return int(_BOOT_OFFSET + (time.time() - _PROCESS_START))
def boot_epoch() -> int:
"""Fake wall-clock boot time for this instance (seconds since epoch)."""
return int(time.time() - uptime_seconds())
def instance_uuid(namespace: str = "") -> str:
"""Deterministic UUID4-looking value for this instance+namespace."""
ns = uuid.UUID("00000000-0000-0000-0000-000000000000")
return str(uuid.uuid5(ns, f"{_HOSTNAME}:{namespace}"))
def instance_hex(nbytes: int, namespace: str = "") -> str:
"""Deterministic hex token of given byte length."""
material = f"{_HOSTNAME}:{namespace}".encode()
digest = hashlib.sha256(material).digest()
while len(digest) < nbytes:
digest += hashlib.sha256(digest).digest()
return digest[:nbytes].hex()
def pick(choices: Sequence[T]) -> T:
"""Deterministic choice from a sequence."""
return rng.choice(list(choices))
def pick_weighted(choices: Sequence[tuple[T, float]]) -> T:
"""Deterministic weighted choice. Input: [(item, weight), ...]."""
total = sum(w for _, w in choices)
r = rng.uniform(0, total)
acc = 0.0
for item, w in choices:
acc += w
if r <= acc:
return item
return choices[-1][0]
def random_bytes(n: int, namespace: str = "") -> bytes:
"""Deterministic per-instance byte string of length n."""
out = bytearray()
i = 0
while len(out) < n:
out.extend(
hashlib.sha256(f"{_HOSTNAME}:{namespace}:{i}".encode()).digest()
)
i += 1
return bytes(out[:n])
def fresh_bytes(n: int) -> bytes:
"""Non-deterministic random bytes — for per-connection nonces/salts."""
return os.urandom(n)
async def jitter(min_ms: int = 5, max_ms: int = 120) -> None:
"""Async response-time jitter. Uses unseeded RNG so timing varies
across connections to the same decky — seeded jitter would leak
predictability."""
await asyncio.sleep(random.uniform(min_ms, max_ms) / 1000.0)
def jitter_sync(min_ms: int = 5, max_ms: int = 120) -> None:
"""Blocking jitter for non-asyncio servers."""
time.sleep(random.uniform(min_ms, max_ms) / 1000.0)

View File

@@ -0,0 +1,488 @@
#!/usr/bin/env python3
"""
POP3 server (port 110).
Full POP3 state machine with bait mailbox.
States: AUTHORIZATION → TRANSACTION
Credentials via IMAP_USERS env var (shared with IMAP service).
10 bait emails containing AWS keys, DB passwords, tokens etc.
"""
import asyncio
import os
import time
from pathlib import Path
from syslog_bridge import (
SEVERITY_WARNING,
encode_secret,
forward_syslog,
syslog_line,
write_syslog_file,
)
NODE_NAME = os.environ.get("NODE_NAME", "mailserver")
SERVICE_NAME = "pop3"
LOG_TARGET = os.environ.get("LOG_TARGET", "")
PORT = int(os.environ.get("PORT", "110"))
POP3_BANNER = os.environ.get("POP3_BANNER", f"+OK {NODE_NAME} Dovecot POP3 ready.")
_RAW_USERS = os.environ.get("IMAP_USERS", "admin:admin123,root:toor,mail:mail,user:user")
VALID_USERS: dict[str, str] = {
u: p for part in _RAW_USERS.split(",") if ":" in part for u, p in [part.split(":", 1)]
}
# Path to a directory of ``*.eml`` files dropped by the orchestrator
# emailgen worker (``/var/spool/decnet-emails/`` by convention). When
# set and populated, those EMLs replace the hardcoded fallback list
# below — same semantics as the IMAP template. Empty / missing falls
# back so a fresh deployment is never silent.
_EMAIL_SEED_PATH = os.environ.get("POP3_EMAIL_SEED", "")
_SEED_RESCAN_INTERVAL = float(os.environ.get("POP3_EMAIL_SEED_RESCAN", "5"))
# ── Bait emails ───────────────────────────────────────────────────────────────
_BAIT_EMAILS: list[str] = [
(
"Date: Mon, 06 Nov 2023 09:12:33 +0000\r\n"
"From: DevOps Team <devops@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: AWS credentials rotation\r\n"
"Message-ID: <1@company.internal>\r\n"
"\r\n"
"Team,\r\n\r\n"
"New AWS credentials have been issued. Old keys deactivated.\r\n\r\n"
"Access Key ID: AKIAIOSFODNN7EXAMPLE\r\n"
"Secret Access Key: wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY\r\n\r\n"
"Update ~/.aws/credentials immediately.\r\n\r\n-- DevOps\r\n"
),
(
"Date: Tue, 07 Nov 2023 14:05:11 +0000\r\n"
"From: Monitoring <monitoring@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: DB password changed\r\n"
"Message-ID: <2@company.internal>\r\n"
"\r\n"
"Production database password was rotated.\r\n\r\n"
"Connection string: mysql://admin:Sup3rS3cr3t!@10.0.1.5:3306/production\r\n\r\n"
"Update all app configs.\r\n"
),
(
"Date: Wed, 08 Nov 2023 08:30:00 +0000\r\n"
"From: GitHub <noreply@github.com>\r\n"
"To: admin@company.internal\r\n"
"Subject: Your personal access token\r\n"
"Message-ID: <3@company.internal>\r\n"
"\r\n"
"Hi admin,\r\n\r\n"
"A new personal access token was created for your account.\r\n\r\n"
"Token: ghp_16C7e42F292c6912E7710c838347Ae178B4a\r\n\r\n"
"If this wasn't you, revoke it immediately at github.com/settings/tokens.\r\n"
),
(
"Date: Thu, 09 Nov 2023 11:22:47 +0000\r\n"
"From: IT Admin <admin@company.internal>\r\n"
"To: team@company.internal\r\n"
"Subject: VPN config attached\r\n"
"Message-ID: <4@company.internal>\r\n"
"\r\n"
"VPN access details for new starters:\r\n\r\n"
" Host: vpn.company.internal:1194\r\n"
" Protocol: UDP\r\n"
" Username: vpnadmin\r\n"
" Password: VpnP@ss2024\r\n\r\n"
"Config file sent separately via secure channel.\r\n"
),
(
"Date: Fri, 10 Nov 2023 16:45:00 +0000\r\n"
"From: SysAdmin <sysadmin@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: Root password\r\n"
"Message-ID: <5@company.internal>\r\n"
"\r\n"
"New root password for prod servers:\r\n\r\n"
" r00tM3T00!\r\n\r\n"
"Change after first login. Do NOT forward this email.\r\n"
),
(
"Date: Sat, 11 Nov 2023 03:12:04 +0000\r\n"
"From: Backup System <backup@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: Backup job failed\r\n"
"Message-ID: <6@company.internal>\r\n"
"\r\n"
"Nightly backup to 192.168.1.50:/mnt/nas FAILED at 03:11 UTC.\r\n\r\n"
"Error: Authentication failed. Credentials in /etc/backup.conf may be stale.\r\n\r\n"
"Last successful backup: 2023-11-10 03:11 UTC\r\n"
),
(
"Date: Sun, 12 Nov 2023 07:04:31 +0000\r\n"
"From: Security Alerts <alerts@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: SSH brute-force alert\r\n"
"Message-ID: <7@company.internal>\r\n"
"\r\n"
"47 failed SSH login attempts detected against prod-web-01.\r\n\r\n"
"Source IPs: 185.220.101.34, 185.220.101.47, 185.220.101.52\r\n"
"Target user: root\r\n"
"Period: 2023-11-12 06:58 - 07:04 UTC\r\n\r\n"
"All attempts blocked by fail2ban. No successful logins.\r\n"
),
(
"Date: Mon, 13 Nov 2023 10:11:55 +0000\r\n"
"From: External Vendor <vendor@external.com>\r\n"
"To: admin@company.internal\r\n"
"Subject: RE: API integration\r\n"
"Message-ID: <8@company.internal>\r\n"
"\r\n"
"Hi,\r\n\r\n"
"Here is the live API key for the integration:\r\n\r\n"
" sk_live_9mK3xF2aP7qR1bN8cT4dW6vE0yU5hJ\r\n\r\n"
"Keep this confidential. Let me know if you need the webhook secret.\r\n\r\n"
"Best regards,\r\nVendor Support\r\n"
),
(
"Date: Tue, 14 Nov 2023 13:48:22 +0000\r\n"
"From: Help Desk <helpdesk@company.internal>\r\n"
"To: admin@company.internal\r\n"
"Subject: Password reset request\r\n"
"Message-ID: <9@company.internal>\r\n"
"\r\n"
"Hi,\r\n\r\n"
"Could you reset my MFA? Current password is Winter2024! so you can verify it's me.\r\n\r\n"
"Thanks\r\n"
),
(
"Date: Wed, 15 Nov 2023 00:01:00 +0000\r\n"
"From: AWS Billing <noreply@aws.amazon.com>\r\n"
"To: admin@company.internal\r\n"
"Subject: Your AWS bill is ready\r\n"
"Message-ID: <10@company.internal>\r\n"
"\r\n"
"Your AWS bill for October 2023 is $847.23.\r\n\r\n"
"Top services:\r\n"
" EC2 (us-east-1): $412.10\r\n"
" RDS (us-east-1): $198.50\r\n"
" S3: $87.43\r\n"
" EC2 (eu-west-2): $149.20\r\n\r\n"
"Account ID: 123456789012\r\n"
),
]
# ── Spool-backed email loader ─────────────────────────────────────────────────
# POP3 stores each message as a single str (full RFC 822 text); when the
# emailgen spool is configured, we read every *.eml in it and serve the
# raw bytes as the corpus. Same caching strategy as the IMAP template.
_seed_cache: list[str] | None = None
_seed_cache_dir_mtime: float = 0.0
_seed_cache_loaded_at: float = 0.0
def _scan_seed_dir(path: Path) -> list[str]:
"""Walk *path* recursively and return each .eml's raw text content,
sorted by mtime so older threads get lower indices."""
eml_paths: list[Path] = []
try:
for p in path.rglob("*.eml"):
if p.is_file():
eml_paths.append(p)
except OSError:
return []
eml_paths.sort(key=lambda p: p.stat().st_mtime)
out: list[str] = []
for p in eml_paths:
try:
out.append(p.read_text(encoding="utf-8", errors="replace"))
except OSError:
continue
return out
def _get_emails() -> list[str]:
"""Return the active corpus. Same fallback rules as IMAP template."""
global _seed_cache, _seed_cache_dir_mtime, _seed_cache_loaded_at
if not _EMAIL_SEED_PATH:
return _BAIT_EMAILS
seed_dir = Path(_EMAIL_SEED_PATH)
try:
dir_stat = seed_dir.stat()
except OSError:
return _BAIT_EMAILS
now = time.monotonic()
fresh_enough = (
_seed_cache is not None
and (now - _seed_cache_loaded_at) < _SEED_RESCAN_INTERVAL
and dir_stat.st_mtime == _seed_cache_dir_mtime
)
if fresh_enough:
return _seed_cache or _BAIT_EMAILS
scanned = _scan_seed_dir(seed_dir)
if not scanned:
return _BAIT_EMAILS
_seed_cache = scanned
_seed_cache_dir_mtime = dir_stat.st_mtime
_seed_cache_loaded_at = now
return scanned
# ── Logging ───────────────────────────────────────────────────────────────────
def _log(event_type: str, severity: int = 6, **kwargs) -> None:
line = syslog_line(SERVICE_NAME, NODE_NAME, event_type, severity, **kwargs)
write_syslog_file(line)
forward_syslog(line, LOG_TARGET)
# ── Protocol ──────────────────────────────────────────────────────────────────
class POP3Protocol(asyncio.Protocol):
def __init__(self):
self._transport = None
self._peer = ("?", 0)
self._buf = b""
self._state = "AUTHORIZATION"
self._current_user: str | None = None
self._deleted: set[int] = set() # 0-based indices of DELE'd messages
def connection_made(self, transport):
self._transport = transport
self._peer = transport.get_extra_info("peername", ("?", 0))
_log("connect", src=self._peer[0], src_port=self._peer[1])
banner = POP3_BANNER if POP3_BANNER.endswith("\r\n") else POP3_BANNER + "\r\n"
if not banner.startswith("+OK"):
banner = "+OK " + banner
transport.write(banner.encode())
def data_received(self, data):
self._buf += data
while b"\n" in self._buf:
line, self._buf = self._buf.split(b"\n", 1)
self._handle_line(line.decode(errors="replace").strip())
def connection_lost(self, exc):
_log("disconnect", src=self._peer[0] if self._peer else "?")
# ── Command dispatch ──────────────────────────────────────────────────────
def _handle_line(self, line: str) -> None:
parts = line.split(None, 1)
if not parts:
return
cmd = parts[0].upper()
args = parts[1] if len(parts) > 1 else ""
_log("command", src=self._peer[0], cmd=cmd, state=self._state)
# Always available
if cmd == "CAPA":
self._transport.write(
b"+OK\r\nTOP\r\nUSER\r\nUIDL\r\nRESP-CODES\r\nAUTH-RESP-CODE\r\nSASL\r\n.\r\n"
)
elif cmd == "QUIT":
self._transport.write(b"+OK Logging out.\r\n")
self._transport.close()
# AUTHORIZATION state
elif cmd == "USER":
self._cmd_user(args)
elif cmd == "PASS":
self._cmd_pass(args)
# TRANSACTION state
elif cmd == "STAT":
self._cmd_stat()
elif cmd == "LIST":
self._cmd_list(args)
elif cmd == "RETR":
self._cmd_retr(args)
elif cmd == "TOP":
self._cmd_top(args)
elif cmd == "UIDL":
self._cmd_uidl(args)
elif cmd == "DELE":
self._cmd_dele(args)
elif cmd == "RSET":
self._cmd_rset()
elif cmd == "NOOP":
self._transport.write(b"+OK\r\n")
else:
self._transport.write(b"-ERR Command not recognized\r\n")
# ── Command implementations ───────────────────────────────────────────────
def _cmd_user(self, args: str) -> None:
if self._state != "AUTHORIZATION":
self._transport.write(b"-ERR Already authenticated\r\n")
return
self._current_user = args.strip()
self._transport.write(b"+OK User name accepted, password please\r\n")
def _cmd_pass(self, args: str) -> None:
if self._state != "AUTHORIZATION":
self._transport.write(b"-ERR Already authenticated\r\n")
return
if not self._current_user:
self._transport.write(b"-ERR USER required first\r\n")
return
username = self._current_user
password = args.strip()
_enc = encode_secret(password)
if VALID_USERS.get(username) == password:
self._state = "TRANSACTION"
_log("auth", src=self._peer[0], username=username, principal=username,
outcome="success", **_enc)
self._transport.write(b"+OK Logged in.\r\n")
else:
_log("auth", src=self._peer[0], username=username, principal=username,
outcome="failure", severity=SEVERITY_WARNING, **_enc)
self._current_user = None
self._transport.write(b"-ERR Authentication failed.\r\n")
def _require_transaction(self) -> bool:
if self._state != "TRANSACTION":
self._transport.write(b"-ERR Not authenticated\r\n")
return False
return True
def _active_messages(self) -> list[tuple[int, str]]:
"""Return [(1-based-num, body), ...] excluding DELE'd messages."""
return [
(i + 1, body)
for i, body in enumerate(_get_emails())
if i not in self._deleted
]
def _cmd_stat(self) -> None:
if not self._require_transaction():
return
msgs = self._active_messages()
total = sum(len(b.encode()) for _, b in msgs)
self._transport.write(f"+OK {len(msgs)} {total}\r\n".encode())
def _cmd_list(self, args: str) -> None:
if not self._require_transaction():
return
emails = _get_emails()
if args:
try:
n = int(args)
idx = n - 1
if idx in self._deleted or not (0 <= idx < len(emails)):
self._transport.write(b"-ERR No such message\r\n")
else:
size = len(emails[idx].encode())
self._transport.write(f"+OK {n} {size}\r\n".encode())
except ValueError:
self._transport.write(b"-ERR Invalid argument\r\n")
else:
msgs = self._active_messages()
total = sum(len(b.encode()) for _, b in msgs)
self._transport.write(f"+OK {len(msgs)} messages ({total} octets)\r\n".encode())
for n, body in msgs:
self._transport.write(f"{n} {len(body.encode())}\r\n".encode())
self._transport.write(b".\r\n")
def _cmd_retr(self, args: str) -> None:
if not self._require_transaction():
return
try:
n = int(args)
idx = n - 1
emails = _get_emails()
if idx in self._deleted or not (0 <= idx < len(emails)):
self._transport.write(b"-ERR No such message\r\n")
return
body = emails[idx]
raw = body.encode()
_log("retr", src=self._peer[0], message_num=n)
self._transport.write(f"+OK {len(raw)} octets\r\n".encode())
self._transport.write(raw)
if not raw.endswith(b"\r\n"):
self._transport.write(b"\r\n")
self._transport.write(b".\r\n")
except ValueError:
self._transport.write(b"-ERR Invalid argument\r\n")
def _cmd_top(self, args: str) -> None:
if not self._require_transaction():
return
try:
parts = args.split(None, 1)
n = int(parts[0])
line_count = int(parts[1]) if len(parts) > 1 else 0
idx = n - 1
emails = _get_emails()
if idx in self._deleted or not (0 <= idx < len(emails)):
self._transport.write(b"-ERR No such message\r\n")
return
body = emails[idx]
sep = "\r\n\r\n"
if sep in body:
headers, rest = body.split(sep, 1)
headers += sep
else:
headers, rest = body, ""
body_lines = rest.split("\r\n")[:line_count]
result = headers + "\r\n".join(body_lines)
self._transport.write(b"+OK\r\n")
self._transport.write(result.encode())
if not result.endswith("\r\n"):
self._transport.write(b"\r\n")
self._transport.write(b".\r\n")
except (ValueError, IndexError):
self._transport.write(b"-ERR Invalid arguments\r\n")
def _cmd_uidl(self, args: str) -> None:
if not self._require_transaction():
return
if args:
try:
n = int(args)
idx = n - 1
if idx in self._deleted or not (0 <= idx < len(_get_emails())):
self._transport.write(b"-ERR No such message\r\n")
else:
self._transport.write(f"+OK {n} msg-{n}\r\n".encode())
except ValueError:
self._transport.write(b"-ERR Invalid argument\r\n")
else:
self._transport.write(b"+OK\r\n")
for n, _ in self._active_messages():
self._transport.write(f"{n} msg-{n}\r\n".encode())
self._transport.write(b".\r\n")
def _cmd_dele(self, args: str) -> None:
if not self._require_transaction():
return
try:
n = int(args)
idx = n - 1
if idx in self._deleted or not (0 <= idx < len(_get_emails())):
self._transport.write(b"-ERR No such message\r\n")
else:
self._deleted.add(idx)
_log("delete", src=self._peer[0], message_num=n)
self._transport.write(f"+OK Message {n} deleted\r\n".encode())
except ValueError:
self._transport.write(b"-ERR Invalid argument\r\n")
def _cmd_rset(self) -> None:
if not self._require_transaction():
return
self._deleted.clear()
self._transport.write(b"+OK\r\n")
async def main():
_log("startup", msg=f"POP3 server starting as {NODE_NAME}")
loop = asyncio.get_running_loop()
server = await loop.create_server(POP3Protocol, "0.0.0.0", PORT) # nosec B104
async with server:
await server.serve_forever()
if __name__ == "__main__":
asyncio.run(main())

View File

@@ -0,0 +1,261 @@
#!/usr/bin/env python3
"""
Shared RFC 5424 syslog helper used by service containers.
Services call syslog_line() to format an RFC 5424 message, then
write_syslog_file() to emit it to stdout — the container runtime
captures it, and the host-side collector streams it into the log file.
RFC 5424 structure:
<PRI>1 TIMESTAMP HOSTNAME APP-NAME PROCID MSGID [SD-ELEMENT] MSG
Facility: local0 (16). SD element ID uses PEN 55555.
"""
import base64
import re
from datetime import datetime, timezone
from typing import Any, Optional
# ─── Constants ────────────────────────────────────────────────────────────────
_FACILITY_LOCAL0 = 16
_SD_ID = "relay@55555"
_NILVALUE = "-"
SEVERITY_EMERG = 0
SEVERITY_ALERT = 1
SEVERITY_CRIT = 2
SEVERITY_ERROR = 3
SEVERITY_WARNING = 4
SEVERITY_NOTICE = 5
SEVERITY_INFO = 6
SEVERITY_DEBUG = 7
_MAX_HOSTNAME = 255
_MAX_APPNAME = 48
_MAX_MSGID = 32
# ─── Formatter ────────────────────────────────────────────────────────────────
def _sd_escape(value: str) -> str:
"""Escape SD-PARAM-VALUE per RFC 5424 §6.3.3."""
return value.replace("\\", "\\\\").replace('"', '\\"').replace("]", "\\]")
def _sd_element(fields: dict[str, Any]) -> str:
if not fields:
return _NILVALUE
params = " ".join(f'{k}="{_sd_escape(str(v))}"' for k, v in fields.items())
return f"[{_SD_ID} {params}]"
def syslog_line(
service: str,
hostname: str,
event_type: str,
severity: int = SEVERITY_INFO,
timestamp: datetime | None = None,
msg: str | None = None,
**fields: Any,
) -> str:
"""
Return a single RFC 5424-compliant syslog line (no trailing newline).
Args:
service: APP-NAME (e.g. "http", "mysql")
hostname: HOSTNAME (node name)
event_type: MSGID (e.g. "request", "login_attempt")
severity: Syslog severity integer (default: INFO=6)
timestamp: UTC datetime; defaults to now
msg: Optional free-text MSG
**fields: Encoded as structured data params
"""
pri = f"<{_FACILITY_LOCAL0 * 8 + severity}>"
ts = (timestamp or datetime.now(timezone.utc)).isoformat()
host = (hostname or _NILVALUE)[:_MAX_HOSTNAME]
appname = (service or _NILVALUE)[:_MAX_APPNAME]
msgid = (event_type or _NILVALUE)[:_MAX_MSGID]
sd = _sd_element(fields)
message = f" {msg}" if msg else ""
return f"{pri}1 {ts} {host} {appname} {_NILVALUE} {msgid} {sd}{message}"
def encode_secret(secret: str) -> dict[str, str]:
"""Standardized credential-secret encoding for the universal SD-block shape.
Returns ``{'secret_printable': ..., 'secret_b64': ...}`` ready to spread
into a :func:`syslog_line` / ``_log`` call::
_log("auth_attempt", principal=user, **encode_secret(password))
``secret_printable`` mirrors auth-helper.c's sd_escape: bytes outside
``[0x20, 0x7f)`` collapse to ``'?'`` so the field is always parser-safe
RFC 5424 ASCII. ``secret_b64`` preserves the *original* utf-8 bytes —
NUL/0xff/control/non-utf8 sequences all survive losslessly, useful as
a fingerprinting signal even when the printable form sanitizes them.
The decnet web ingester's native-shape branch keys off ``secret_b64``
being present, so any service emitter calling this helper lands its
cred attempt directly in the :class:`Credential` table.
"""
raw = secret.encode("utf-8", errors="replace")
printable = "".join(chr(b) if 0x20 <= b < 0x7f else "?" for b in raw)
return {
"secret_printable": printable,
"secret_b64": base64.b64encode(raw).decode("ascii"),
}
_DIGEST_PARAM_RE = re.compile(r'(\w+)\s*=\s*"([^"]*)"|(\w+)\s*=\s*([^,\s]+)')
def classify_authorization(header_value: Optional[str]) -> Optional[dict[str, Any]]:
"""Parse an HTTP Authorization header value into Credential SD fields.
Returns a dict with the universal cred shape ready to spread into a
``_log(...)`` call::
auth = request.headers.get("Authorization")
cred = classify_authorization(auth)
if cred:
_log("auth_attempt", **cred)
Recognised schemes:
* Basic — base64(user:pw); decoded → ``principal=user`` +
``secret_kind="plaintext"`` + ``encode_secret(pw)``.
* Bearer / Token — opaque token; ``principal=None`` +
``secret_kind="http_bearer"`` + ``encode_secret(token)``.
* Digest — ``principal=username`` from header +
``secret_kind="http_digest_md5"`` + ``encode_secret(response)``.
Returns ``None`` for anything unrecognized (AWS4-HMAC-SHA256, NTLM,
Negotiate, …) — callers can still log the raw header value in the
ambient SD-block; we just don't know how to extract a hashable
secret from it.
"""
if not header_value or not isinstance(header_value, str):
return None
parts = header_value.strip().split(None, 1)
if len(parts) < 2:
return None
scheme, rest = parts[0].lower(), parts[1].strip()
if scheme == "basic":
try:
decoded = base64.b64decode(rest, validate=True).decode("utf-8", errors="replace")
except (ValueError, base64.binascii.Error):
return None
if ":" not in decoded:
return None
user, _, pw = decoded.partition(":")
return {
"principal": user,
"secret_kind": "plaintext",
**encode_secret(pw),
}
if scheme in ("bearer", "token"):
return {
"principal": None,
"secret_kind": "http_bearer",
**encode_secret(rest),
}
if scheme == "digest":
params: dict[str, str] = {}
for m in _DIGEST_PARAM_RE.finditer(rest):
k = m.group(1) or m.group(3)
v = m.group(2) if m.group(2) is not None else m.group(4)
if k:
params[k.lower()] = v
response = params.get("response")
if not response:
return None
return {
"principal": params.get("username"),
"secret_kind": "http_digest_md5",
**encode_secret(response),
}
return None
_FORM_PRINCIPAL_KEYS = (
"username", "user", "email", "login", "userid", "account",
"log", # wp-login.php
"user_login", # WordPress alt
"uname", # phpMyAdmin
"pma_username",
)
_FORM_SECRET_KEYS = (
"password", "pass", "pwd", "passwd", "passwort", "mot_de_passe",
"user_password", # WordPress alt
"pma_password", # phpMyAdmin
)
def extract_form_credentials(
body: Optional[str],
content_type: Optional[str],
) -> Optional[dict[str, Any]]:
"""Parse an `application/x-www-form-urlencoded` body for credentials.
Returns the universal cred SD shape ready to spread into a
``_log(...)`` call when both a principal-shaped key and a secret-
shaped key are present in the body. Otherwise returns ``None``.
Field-name detection is case-insensitive and covers the most common
login-form variants (WordPress wp-login.php, phpMyAdmin, Joomla,
etc.). Add more entries to ``_FORM_PRINCIPAL_KEYS`` /
``_FORM_SECRET_KEYS`` as new templates surface them.
"""
if not body or not isinstance(content_type, str):
return None
if not content_type.lower().startswith("application/x-www-form-urlencoded"):
return None
fields: dict[str, str] = {}
for pair in body.split("&"):
if "=" not in pair:
continue
k, _, v = pair.partition("=")
# urllib decode without importing urllib at module scope (the
# template emitters are import-cost-sensitive). Inline the
# tiny percent-decode + plus-decode.
try:
from urllib.parse import unquote_plus
key = unquote_plus(k).lower()
val = unquote_plus(v)
except Exception:
continue
# First-wins so duplicate-key forms don't get clobbered.
fields.setdefault(key, val)
principal: Optional[str] = None
for k in _FORM_PRINCIPAL_KEYS:
if k in fields:
principal = fields[k]
break
secret: Optional[str] = None
for k in _FORM_SECRET_KEYS:
if k in fields:
secret = fields[k]
break
if secret is None:
return None
return {
"principal": principal,
"secret_kind": "plaintext",
**encode_secret(secret),
}
def write_syslog_file(line: str) -> None:
"""Emit a syslog line to stdout for container log capture."""
print(line, flush=True)
def forward_syslog(line: str, log_target: str) -> None:
"""No-op stub. TCP forwarding is handled by rsyslog, not by service containers."""
pass