Adding a new server to your inventory shouldn't mean manually typing in CPU model, RAM capacity, serial number, NIC MAC addresses, and storage configuration. And onboarding a new switch shouldn't require hand-counting ports and copying firmware versions from a CLI session. Hardware auto-discovery solves this by pulling specs directly from the devices themselves.
This guide explains the three protocols used for hardware and port discovery in datacenter environments — SNMP, IPMI, and Redfish — and how FluxBilling uses them to auto-populate your inventory.
The Three Discovery Protocols
SNMP: Discovering Network Device Ports
Simple Network Management Protocol (SNMP) is the standard for querying network devices. Every managed switch, router, and firewall supports it. FluxBilling uses SNMP to discover and monitor port information on your network devices.
When you point FluxBilling at a network device with SNMP credentials, it walks the device's MIB (Management Information Base) tables and extracts:
- Interface list — every port on the device with its name (e.g., "Gi1/0/1", "eth0"), description, and alias
- MAC addresses — physical address per interface
- Port speeds — line rate in Mbps (using 64-bit high-speed counters when available)
- Operational status — whether each port is up, down, dormant, or in testing mode
- System identity — hostname, system description, vendor OID
- Firmware version — parsed from the system description string
- Serial number — queried from Entity MIB or vendor-specific OIDs
FluxBilling supports both SNMPv2c (community string) and SNMPv3 (username with SHA/MD5 authentication and AES/DES encryption). All credentials are encrypted at rest.
Vendor Auto-Detection
Different switch vendors format their SNMP data differently. FluxBilling automatically detects the vendor from the system description and adjusts its parsing accordingly:
- Arista — detects "EOS version" in the system description
- Cisco — detects "IOS version"
- Juniper — detects "JUNOS" version strings
- MikroTik — detects "RouterOS" and queries vendor-specific serial number OIDs
- Ubiquiti — detects "EdgeOS" or "UniFi" identifiers
This means you don't need to configure vendor-specific settings. Point FluxBilling at any supported switch, and it figures out the rest.
Smart Interface Filtering
Not every interface on a network device is a physical port you care about. FluxBilling automatically filters out loopback interfaces, null interfaces, VLAN interfaces, bridge interfaces, and tunnel interfaces (GRE, L2TP, PPPoE, PPTP, EoIP, OpenVPN, SSTP, IPIP, 6to4). What you see is a clean list of physical ports and their real-time status.
IPMI: Discovering Server Hardware
Intelligent Platform Management Interface (IPMI) is the standard for out-of-band server management. Every enterprise server has a BMC (Baseboard Management Controller) that exposes hardware information and remote control capabilities.
FluxBilling queries IPMI to discover:
- System info — serial number, model, firmware version from the FRU (Field Replaceable Unit) data
- CPU and memory — processor model, core count, memory capacity
- Network interfaces — MAC addresses and interface names
- Storage — disk count, capacity, and model
- Sensor data — temperatures, fan speeds, power supply status
Redfish: Modern Server Discovery
Redfish is the modern replacement for IPMI, offering a RESTful API over HTTPS. Newer BMCs (Dell iDRAC9+, Supermicro X10+, HPE iLO 5+) support Redfish alongside IPMI.
FluxBilling supports Redfish discovery as a primary method, falling back to IPMI for older hardware. Redfish provides the same hardware data — CPU, memory, storage, NICs — but through a structured JSON API that's more reliable and detailed than IPMI's FRU parsing.
How Auto-Discovery Works in Practice
Single Device Fetch
When adding a new device to inventory, you provide the management IP and credentials. FluxBilling reaches out to the device, queries its hardware specs, and auto-fills the inventory record. For a server, this means CPU, RAM, storage, serial number, and NICs are populated automatically. For a network device, you get the full port list, firmware version, and system identity.
Bulk Discovery
When onboarding an entire rack or a batch of new switches, FluxBilling supports bulk hardware fetch as a background job. You select multiple devices and trigger a batch discovery. The platform queries each device in parallel, populates the results, and reports any failures — so you can fix credentials or connectivity issues without redoing the entire batch.
Agent-Proxied Discovery
Not all devices are directly reachable from your management server. Servers and switches on private networks — behind NAT or in remote locations — need a different path. FluxBilling supports provisioning agents that run in remote locations. When a device has an assigned agent, SNMP and IPMI queries are proxied through that agent, which runs the discovery locally and returns the results to the master platform.
Continuous Port Monitoring
Discovery is just the beginning. Once your network devices are registered with SNMP credentials, FluxBilling polls them every 5 minutes to collect real-time port metrics:
- Traffic rates — bytes per second in and out, calculated from 64-bit counter deltas with wrap detection
- Utilization — percentage of line rate being used per port
- Error and discard counts — indicators of network problems
- Packet rates — unicast packets per second
Metrics are aggregated hourly and cleaned up daily, giving you historical traffic data without unbounded storage growth. This data feeds into traffic charts, capacity planning, and 95th percentile bandwidth billing.
Device Health Monitoring
Beyond port traffic, SNMP polling also collects device health metrics: CPU usage, memory utilization, temperature readings, fan and power supply status, and system uptime. These are stored per-device and used for alerting when thresholds are crossed.
Port Connection Mapping
Knowing what ports exist on your switches is useful. Knowing what's connected to each port is essential. FluxBilling's port connection system maps switch ports to server NICs, creating a network topology that shows exactly how your infrastructure is wired.
You can see which server NIC is connected to which switch port (including VLAN assignments), identify unconnected ports on your switches, find available NICs on your servers, and trace physical connectivity from patch panel to server. For providers using patch panels, FluxBilling tracks front and rear connections with cable type, color, and full path tracing.
Security Considerations
Hardware discovery involves sensitive credentials. FluxBilling addresses this with:
- AES-256-GCM encryption for all stored SNMP community strings, IPMI passwords, and SSH credentials
- SSRF prevention — blocks IPMI/Redfish requests to private/reserved IP ranges that could be used for server-side request forgery
- On-demand decryption — credentials are only decrypted when an authorized admin explicitly requests them
- Timeout protection — all hardware queries have a 60-second timeout with abort controllers to prevent hung connections
Getting Started
If you're manually typing hardware specs into your inventory or running discovery scripts outside your billing platform, integrating auto-discovery eliminates a significant source of errors and wasted time. One fetch populates an entire server's specs. One SNMP walk maps every port on a switch.
FluxBilling includes SNMP, IPMI, and Redfish discovery in every plan — no add-ons or separate licenses. Explore the full feature set to see how hardware discovery fits into the broader platform.
