v0.38.1 · Apache-2.0 · Go 1.27 ·

A network load target
for the monitoring tools
you're building.

nl6 simulates up to 30,000 network devices, GPU servers, storage systems and Linux hosts on a single Linux host — each with its own IP, SNMP listener, SSH server, HTTPS REST endpoint and flow exporter. Built on TUN interfaces and network namespaces.

quick start →github ↗

noc@laptop — ~ — 132×36

noc@laptop:~$

devices / host

30,000max

device types

28in 8 cat.

mem / device

~1KB

parallel workers

500max

quick start

build from source · install a package · or pull with docker

01 · clonegit

$git clone https://github.com/labmonkeys-space/nl6.git

$cd nl6

02 · buildmake · go 1.26+

$make tidy

$make build

03 · runneeds root

$sudo ./go/nl6/nl6 -auto-start-ip 10.0.0.1 -auto-count 100

or install a package

01 · download & install.deb / .rpm · from releases

$curl -LO https://github.com/labmonkeys-space/nl6/releases/download/v0.38.1/nl6_0.38.1_amd64.deb

$sudo apt install ./nl6_0.38.1_amd64.deb

$curl -LO https://github.com/labmonkeys-space/nl6/releases/download/v0.38.1/nl6-0.38.1-1.x86_64.rpm

$sudo dnf install ./nl6-0.38.1-1.x86_64.rpm

02 · configure & startsystemd · needs root

$sudoedit /etc/nl6/nl6.conf

$sudo systemctl enable --now nl6

or on nixos · declarative flake

01 · enable the moduleservices.nl6

#services.nl6.enable = true;

02 · prebuilt cachenix.settings · no cli

#substituters = [ "https://nl6.cachix.org" ];

#trusted-public-keys = [ "nl6.cachix.org-1:nfaq8JEbMcARjzc/oPyNIrcQrXKe13phUtMg0RucnLA=" ];

or with docker

01 · pullno toolchain

$docker pull ghcr.io/labmonkeys-space/nl6:latest

02 · runneeds --cap-add=net_admin

docker run --rm -it \
  --cap-add=NET_ADMIN \
  --device=/dev/net/tun \
  --network=host \
  ghcr.io/labmonkeys-space/nl6:latest \
  -auto-start-ip 10.0.0.1 -auto-count 100

what's in the box

six pillars

30,000 devices

Tested scale on a single host. Parallel TUN pre-allocation, lock-free sync.Map for O(1) OID lookups, pre-computed next-OID mappings.

Protocols

SNMP v1/v2c/v3 (MD5/SHA1 · DES/AES128) with opt-in SET, SSH with VT100, HTTPS REST, gNMI (OpenConfig plus per-type catalogues), NETCONF on Catalyst 9500 and MX10004, LLDP topology. Push exporters: NetFlow v5 / v9 / IPFIX with NBAR2, sFlow v5 (experimental), SNMP traps/informs, syslog over UDP/TCP/TLS, gNMI dial-out.

Load-test fidelity

Arm → run a bounded push experiment over 7 protocols → get an exact sent-ledger. Diff it against your monitor to localize missed or duplicated telemetry. Fidelity mode, loss localization, run tagging, nl6-reconcile CLI, HTML report.

31 device types

Routers, switches, firewalls, servers, GPU servers (DGX/HGX), storage systems and optical transport, across 9 categories.

GPU simulation

NVIDIA DGX-A100 / H100 / HGX-H200 with per-GPU DCGM OIDs — utilization, VRAM, temp, power, fan, SM/memory clocks.

Namespace isolation

Each device runs in the dedicated nl6sim network namespace with its own TUN interface and IP.

Dynamic metrics

100-point sine-wave cycling for CPU, memory, temperature; full IF-MIB counter set (octets, ucast / mcast / bcast packets, errors, discards) with per-device error scenarios; per-GPU DCGM OIDs.

device catalog

31 types · 9 categories

core routers7

edge routers2

›Juniper MX240 · 24
›NEC IX3315 · 48

dc switches2

›Cisco Nexus 9500 · 64
›Arista 7280R3 · 32

campus switches4

›Cisco Catalyst 9500 · 48
›Extreme VSP4450 · 48
›D-Link DGS-3630 · 52
›Cisco IOS · 4

firewalls4

›Palo Alto PA-3220 · 16
›Fortinet FortiGate-600E · 20
›SonicWall NSa 6700 · 16
›Check Point 15600 · 24

servers4

›Dell PowerEdge R750
›HPE ProLiant DL380
›IBM Power S922
›Linux Server · Ubuntu 24.04

gpu servers3

›NVIDIA DGX-A100 · 8×80GB
›NVIDIA DGX-H100 · 8×80GB
›NVIDIA HGX-H200 · 8×141GB

storage4

›AWS S3
›Pure Storage FlashArray
›NetApp ONTAP
›Dell EMC Unity

optical1

›Ciena Waveserver 5

status & scale

what works · how big

stable12

■SNMP v1/v2c/v3 + SET
■SSH (VT100)
■HTTPS REST (storage)
■NetFlow v5/v9/IPFIX + NBAR2
■gNMI (dial-in + dial-out)
■NETCONF (Catalyst 9500, MX10004)
■SNMP trap + syslog export
■LLDP topology
■DNS + Prometheus service discovery
■Load-test scenarios
■TUN + netns isolation
■Web UI + REST API

experimental1

■sFlow v5 (synthetic)

tested scale3

■30,000 concurrent devices / host
■~50 MB base + ~1 KB / device
■CPU: minimal in steady state

concurrent devices

30,000tested

device types

resource files

341json

world cities

98sysLocation

ssh commands

36+linux

documentation map

jump in

getting started

Install from a package, Ansible or Docker, then start a fleet, resolve it with DNS and enable SNMP SET.

ops

Scale to 30k devices, tune the nl6sim namespace, troubleshoot, profile, and run exports and load tests.

reference

CLI flags, REST API, device data, SNMP, gNMI, NETCONF, export wire formats, optical, discovery and GPU simulation.

explanation

Why it is built this way: architecture, Kubernetes, SNMP data fidelity, collector ceilings.

→ get started

Spin up tens of thousands of devices in seconds.

Apache-2.0. No agents, no cloud, no per-device fees. Just TUN interfaces and a little Go.