SNMP traps

SNMP traps (event-driven)

A trap is an alert a device emits on its own when something noteworthy happens — a link failure, a reboot, a threshold breach. VerOps listens for these, normalizes them into a consistent event model, and enriches them with what it already knows about the device.

Turn it on

1. Enable SNMP on the network agent that sits on the same network as your devices. The agent then listens for traps on the standard SNMP trap port, UDP 162.
2. On each device, add VerOps as a trap destination (a "trap sink" / "notification target"): point it at the agent's IP address on port 162. Set a community string (SNMPv1/v2c) or credentials (SNMPv3) to match.
3. Send a test trap from the device. It appears in Network → SNMP → Traps within seconds.

SNMPv1, v2c and v3 are all fully supported — including v3 authPriv (authenticated and encrypted) traps, which VerOps verifies and decrypts into real events. See SNMPv3 users below to configure the credentials.

SNMPv3 users

To receive authenticated / encrypted (authPriv) v3 traps, add the device's SNMPv3 user under Network → SNMP → SNMPv3 users → Add user. For each user you provide:

• Username — the SNMPv3 user name configured on the device.
• Agent — the agent whose trap listener should recognise this user.
• Authentication — protocol MD5 or SHA plus the auth passphrase.
• Privacy — protocol DES or AES-128 plus the privacy passphrase (for authPriv). Leave privacy empty for authNoPriv.

Keys are stored encrypted and are never shown back — you see only a "configured" indicator. Enable, edit or remove users at any time; with a user in place the agent verifies and decrypts that device's v3 traps automatically.

What you get

• Device registry: every device that has ever sent a trap, with its name, vendor, last-seen time and most recent severity.
• Severity: each trap is classified Critical / Warning / Info from the standard trap definitions (for example, a link-down is Critical).
• Recent-traps stream: a live, filterable feed of incoming traps — filter by device, severity or trap type.
• Top talkers: the noisiest devices and the most frequent trap types over a window, so a flapping interface is obvious at a glance.

Naming your own traps

Standard traps (cold start, link up/down, authentication failure, and so on) are recognized out of the box. For a vendor-specific trap, add a custom trap definition — map its identifier to a friendly name and a severity — and every future occurrence is labelled and coloured accordingly.

Alerting

Actionable traps raise alerts through the same alerting system as the rest of VerOps — there is no separate SNMP alerting to configure. A critical trap (say, an interface going down) opens an alert you can route to your existing channels, with built-in noise control so a storm of repeats is grouped rather than paging you hundreds of times.

Metrics

Traps also produce metrics you can chart and alert on like any other:

Metric	Meaning
snmp_traps_received_total	Traps received.
snmp_traps_failed_total	Traps that could not be processed.
snmp_trap_count	Trap count, broken down by severity.
snmp_trap_rate	Traps per interval — spot a storm.
snmp_device_critical_events	Critical events per device.
snmp_unknown_oid_total	Traps whose type you have not named yet.

Only send authorized devices to your agent. On the device side you are adding a trap destination; on the VerOps side you enable the listener and (optionally) restrict which community string is accepted.