CONNEX gives you total visibility into every network connection, routing table, interface, and protocol — all in real time. No blind spots. No guesswork. Just pure, unfiltered network intelligence.
CONNEX surfaces every connection, every socket, every route — with surgical precision. Here's what ships out of the box.
Zoom into every TCP connection with -t. See established sessions, SYN states,
FIN_WAIT queues — the full lifecycle of every handshake on your infrastructure.
Fire-and-forget protocols deserve first-class observability. Our -u engine maps
every UDP datagram path across your entire fleet.
Instantly surface every listening socket with -l. Know exactly what's open, what's
waiting, and what's exposed — before attackers do.
Enable -a and see everything — listening and non-listening sockets, established
connections, TIME_WAIT remnants. Zero blind spots.
Skip the DNS overhead. -n delivers raw numeric addresses and ports at wire speed.
Sub-millisecond resolution for when every nanosecond counts.
Know exactly which process owns which socket. -p maps PID-to-connection, giving you
the kill chain from process to port to protocol.
Visualize your entire routing topology with -r. Gateways, genmasks, flags,
interfaces — your kernel routing table, surfaced as actionable intelligence.
Deep-dive into per-protocol statistics with -s. TCP segments, UDP packets, IP
forwarding rates — SNMP-level telemetry at your fingertips.
Three steps. No agents. No configuration files. Just plug in and see.
Choose your protocol layer — TCP, UDP, raw sockets, or all of the above. Filter by IPv4
(-4) or IPv6 (-6). Set your address family and go.
Activate -c for real-time continuous monitoring. CONNEX refreshes your
connection state every second, streaming live updates as your network evolves.
Enable -W for panoramic wide output — no truncated addresses. Stack
-e twice for maximum extended detail. Cross-reference with -p for
full process-to-port attribution.
Export interface statistics (-i), routing tables (-r), multicast
groups (-g), and protocol stats (-s) into your observability
stack. Done.
Watch connections form and dissolve in real time. Every node is a socket. Every line is a live connection.
Click any column header to sort. This is what Total Visibility Mode looks like in production.
| Proto ⇅ | Recv-Q ⇅ | Send-Q ⇅ | Local Address ⇅ | Foreign Address ⇅ | State ⇅ | PID/Program ⇅ |
|---|
Per-protocol statistics powered by our -s engine. The insights your network team has
been begging for.
$ connex -tulnp Active Internet connections (only servers) Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 1234/sshd tcp 0 0 127.0.0.1:5432 0.0.0.0:* LISTEN 5678/postgres tcp 0 0 0.0.0.0:80 0.0.0.0:* LISTEN 9012/nginx tcp 0 0 0.0.0.0:443 0.0.0.0:* LISTEN 9012/nginx tcp6 0 0 :::3000 :::* LISTEN 3456/node tcp6 0 0 :::8080 :::* LISTEN 7890/java udp 0 0 0.0.0.0:68 0.0.0.0:* 2345/dhclient udp 0 0 0.0.0.0:123 0.0.0.0:* 6789/ntpd
Every tier unlocks more protocol layers, more connection states, and deeper observability. No hidden fees.
Basic socket visibility for solo developers
-t)-n)Full-spectrum observability for serious operators
-t -u -w)-l -a)-p)-c)-r)-e)Team-wide network intelligence with full audit trails
-i)-s)-g)-W)-4 -6)Dedicated network observability for mission-critical infrastructure
-M)-ee)-o)-Z)"We were flying blind on socket states. CONNEX's Total Visibility Mode gave us the connection-level observability we'd been trying to build internally for months. The Process Attribution feature alone saved our on-call team countless hours of debugging."
"I just run connex -tulnp and I know exactly what's
listening on every port, which process owns it, and whether something unauthorized is open. It's
become the first thing I check in every incident response runbook."
"The Protocol Analytics Dashboard changed how we think about capacity planning. Seeing TCP retransmit rates and UDP packet loss in real time — that's the kind of telemetry that actually moves the needle on SLAs."
"Continuous Stream mode is genuinely addictive. We pipe CONNEX output into our alerting stack and catch connection anomalies within seconds. Route Mapping is the cherry on top — we deprecated three internal tools after adopting it."
CONNEX supports TCP, UDP, UDPLite, SCTP, raw sockets, UNIX domain sockets, Bluetooth (L2CAP
and RFCOMM), IPX, AX.25, NET/ROM, DDP (AppleTalk), and more. Use protocol-specific flags
like --tcp, --udp, --raw, or specify address families
with -A for surgical precision.
Absolutely. Our Process Attribution™ engine (-p) shows the PID and program name
for every socket. You'll need elevated privileges for sockets owned by other users — CONNEX
respects your security model while delivering maximum visibility.
Yes — Continuous Stream™ (-c) prints your selected information every second,
refreshing automatically. Pair it with -tulnp for a live dashboard of every
listening TCP and UDP socket with process attribution. It's the closest thing to a network
heartbeat monitor.
By default, CONNEX optimizes for terminal width. Enable Panoramic View™ (-W) for
full, untruncated addresses — as wide as needed. Pair with -n for raw numeric
output without DNS resolution overhead.
Of course. CONNEX reads directly from your kernel's networking subsystem —
/proc/net/tcp, /proc/net/udp, /proc/net/unix, and
more. No external API calls. No cloud dependency. Your network data never leaves your
machine.