How to Build Full-Stack H3C Device Monitoring with SNMP Exporter, Prometheus & Grafana

The article presents a full monitoring architecture for H3C network devices, illustrated by a diagram that separates the data source, collection, storage/computation, and visualization layers.

Data source layer: Switches expose operational metrics such as port traffic, CPU usage, memory and status via the SNMP protocol.

Collection layer: The SNMP Exporter actively pulls SNMP data from the switches, converts it into Prometheus‑compatible metric format, and exposes it through an HTTP endpoint (default port 9116).

Storage & computation layer: Prometheus scrapes the exporter, stores the time‑series data, and provides aggregation and alert‑rule evaluation capabilities.

Visualization layer: Grafana connects to Prometheus as a data source and builds dashboards that display traffic, status, and other key indicators.

All switches must have SNMP enabled before the exporter can collect data. For H3C devices the configuration steps are:

system-view
snmp-agent
# Configure SNMPv2c read‑only community
snmp-agent community read public
# Specify SNMP Exporter as management host
snmp-agent target-host trap address udp-domain 192.168.1.100 securityname public v2c
save

The SNMP Exporter (version snmp_exporter-0.30.0.linux-amd64.tar.gz) is then installed and its snmp.yaml file is edited to list the metrics to be collected.

After starting the exporter, Prometheus is configured with a scrape job pointing to the exporter’s endpoint. Once Prometheus shows the target as up, the switch performance data can be retrieved via the provided URL.

Finally, Grafana is used to import an SNMP‑related dashboard template from the marketplace. The dashboard is adjusted to display the collected metrics, resulting in a clear visual representation of interface traffic, CPU load, memory usage, and device status.

The end‑to‑end process—from SNMP configuration on the switch to metric visualization in Grafana—is concise, reproducible, and demonstrates how the Prometheus ecosystem can be leveraged for effective network device monitoring.