feat: complete OTEL tracing across all services with pipeline bridge and docs

Extends tracing to every remaining module: all 23 API route handlers,
correlation engine, sniffer (fingerprint/p0f/syslog), prober (jarm/hassh/tcpfp),
profiler behavioral analysis, logging subsystem, engine, and mutator.

Bridges the ingester→SSE trace gap by persisting trace_id/span_id columns on
the logs table and creating OTEL span links in the SSE endpoint. Adds log-trace
correlation via _TraceContextFilter injecting otel_trace_id into Python LogRecords.

Includes development/docs/TRACING.md with full span reference (76 spans),
pipeline propagation architecture, quick start guide, and troubleshooting.
This commit is contained in:
2026-04-16 00:58:08 -04:00
parent 04db13afae
commit 70d8ffc607
38 changed files with 577 additions and 124 deletions

View File

@@ -9,7 +9,7 @@ This is the *server* variant of HASSH (HASSHServer). It fingerprints what
the server *offers*, which identifies the SSH implementation (OpenSSH,
Paramiko, libssh, Cobalt Strike SSH, etc.).
Stdlib only (socket, struct, hashlib). No DECNET imports.
Stdlib only (socket, struct, hashlib) plus decnet.telemetry for tracing (zero-cost when disabled).
"""
from __future__ import annotations
@@ -19,6 +19,8 @@ import socket
import struct
from typing import Any
from decnet.telemetry import traced as _traced
# SSH protocol constants
_SSH_MSG_KEXINIT = 20
_KEX_INIT_COOKIE_LEN = 16
@@ -36,6 +38,7 @@ _MAX_PACKET_LEN = 35000
# ─── SSH connection + KEX_INIT capture ──────────────────────────────────────
@_traced("prober.hassh_ssh_connect")
def _ssh_connect(
host: str,
port: int,
@@ -213,6 +216,7 @@ def _compute_hassh(kex: str, enc: str, mac: str, comp: str) -> str:
# ─── Public API ─────────────────────────────────────────────────────────────
@_traced("prober.hassh_server")
def hassh_server(
host: str,
port: int,

View File

@@ -8,7 +8,7 @@ fingerprint that identifies the TLS server implementation.
Reference: https://github.com/salesforce/jarm
No DECNET imports — this module is self-contained and testable in isolation.
Only DECNET import is decnet.telemetry for tracing (zero-cost when disabled).
"""
from __future__ import annotations
@@ -19,6 +19,8 @@ import struct
import time
from typing import Any
from decnet.telemetry import traced as _traced
# ─── Constants ────────────────────────────────────────────────────────────────
JARM_EMPTY_HASH = "0" * 62
@@ -379,6 +381,7 @@ def _version_to_str(version: int) -> str:
# ─── Probe sender ────────────────────────────────────────────────────────────
@_traced("prober.jarm_send_probe")
def _send_probe(host: str, port: int, hello: bytes, timeout: float = 5.0) -> bytes | None:
"""
Open a TCP connection, send the ClientHello, and read the ServerHello.
@@ -471,6 +474,7 @@ def _compute_jarm(responses: list[str]) -> str:
# ─── Public API ──────────────────────────────────────────────────────────────
@_traced("prober.jarm_hash")
def jarm_hash(host: str, port: int, timeout: float = 5.0) -> str:
"""
Compute the JARM fingerprint for a TLS server.

View File

@@ -15,6 +15,8 @@ import hashlib
import random
from typing import Any
from decnet.telemetry import traced as _traced
# Lazy-import scapy to avoid breaking non-root usage of HASSH/JARM.
# The actual import happens inside functions that need it.
@@ -36,6 +38,7 @@ _OPT_CODES: dict[str, str] = {
# ─── Packet construction ───────────────────────────────────────────────────
@_traced("prober.tcpfp_send_syn")
def _send_syn(
host: str,
port: int,
@@ -196,6 +199,7 @@ def _compute_fingerprint(fields: dict[str, Any]) -> tuple[str, str]:
# ─── Public API ─────────────────────────────────────────────────────────────
@_traced("prober.tcp_fingerprint")
def tcp_fingerprint(
host: str,
port: int,

View File

@@ -169,6 +169,7 @@ def _write_event(
# ─── Target discovery from log stream ────────────────────────────────────────
@_traced("prober.discover_attackers")
def _discover_attackers(json_path: Path, position: int) -> tuple[set[str], int]:
"""
Read new JSON log lines from the given position and extract unique
@@ -399,6 +400,7 @@ def _tcpfp_phase(
# ─── Main worker ─────────────────────────────────────────────────────────────
@_traced("prober.worker")
async def prober_worker(
log_file: str,
interval: int = 300,