Switches, routers, and firewalls make up the backbone of most networks, and they're almost always monitored over SNMP rather than SSH or an agent — see SNMP Monitoring first if you haven't set that up yet.
What's typically monitored
- Interface status — is each port up, down, or administratively disabled
- Traffic — inbound and outbound throughput per interface, useful for spotting saturated links
- Errors and discards — rising error counts on an interface usually point to a cabling, duplex mismatch, or hardware problem before it becomes a full outage
- Device health — CPU and memory utilization on the device itself, not just the traffic passing through it
- Uptime — how long since the device last rebooted, which is often the first thing worth checking during an incident
Getting the full picture
Auto Discovery and the Network Topology Monitoring capability work together here: a discovery scan finds devices and their neighbors, and iPM can build a live topology map showing how everything connects, which is far easier to read during an incident than a flat list of hosts.
Vendor-specific depth
Generic SNMP monitoring works across virtually any network device, but some vendors expose extra detail through their own MIBs that generic checks won't pick up automatically. iPM has more specific guidance for:
- Cisco Device Monitoring
- F5 Device Monitoring
If you're running other network vendors, generic SNMP interface and health checks still apply — the vendor-specific articles are about getting extra depth, not a requirement to monitor at all.