Network Discovery Reference
This page provides a reference for network discovery configuration variables, filter script variables, and discovery sources in NetXMS.
For procedures on configuring network discovery, see Network Discovery.
Server Configuration Variables
Settings saved in Configuration > Network discovery take effect immediately. When changed directly as server configuration variables, the variables marked * take effect only after a server restart.
| Variable | Default | Description |
|---|---|---|
|
1024 |
Block size (number of addresses) for active discovery scans. |
|
1 |
Enable or disable SNMP probing during active network discovery. If enabled, the server sends SNMP requests to detect devices that respond to SNMP but not to ICMP pings. |
|
0 |
Enable or disable TCP probing during active network discovery. If enabled, the server tries to establish TCP connections to a list of well-known ports to detect devices that are not responding to ICMP pings. |
|
0 |
Interval in milliseconds between scanning address blocks during active discovery. |
|
7200 |
Interval in seconds between active network discovery polls. |
|
(empty) |
Schedule used to start active network discovery polls, in cron format. |
|
0 |
Disable probing discovered addresses for NetXMS agent. |
|
0 |
Disable probing discovered addresses for EtherNet/IP support. |
|
0 |
Disable SNMP version 1 when probing discovered addresses for SNMP support. |
|
0 |
Disable SNMP version 2 when probing discovered addresses for SNMP support. |
|
0 |
Disable SNMP version 3 when probing discovered addresses for SNMP support. |
|
0 |
Disable probing discovered addresses for SSH support. |
|
0 |
Discovery filter settings. |
|
0 |
Enable or disable merge of duplicate nodes. When enabled, configuration of duplicate node(s) will be merged into the original node and duplicate(s) will be deleted. |
|
900 |
Interval in seconds between passive network discovery polls. |
|
0 |
Discovery type: 0=disabled, 1=passive only, 2=active only, 3=passive and active. |
|
0 |
Enable or disable the use of the DNS name instead of the IP address as the primary host name of newly discovered nodes. The name is used only if it resolves back to the same address. The object name is not changed. |
|
1 |
Keep the domain part of node names that the configuration poll sets ( |
|
0 |
Use SNMP trap information for new node discovery. |
|
0 |
Use syslog messages for new node discovery. |
Discovery Filter Hooks
Discovery filtering is implemented via two hook scripts. For available variables and attributes, see the NXSL reference:
-
Hook::AcceptNewNode — Stage 1, before host contact (IP/MAC only)
-
Hook::DiscoveryFilter — Stage 2, after capabilities discovered (full
DiscoveredNodeobject)
Discovery Sources
| Source | Description |
|---|---|
ARP tables |
Collected from monitored nodes — read locally via the agent where available, via SNMP as fallback (passive) |
Routing tables |
Collected via SNMP or agent from routers (passive) |
Address range scan |
Active scan of configured address ranges: ICMP ping sweep plus SNMP probing (enabled by default, |
Agent registration |
Agents announcing themselves to the server; controlled by |
SNMP trap source |
Source address of traps from unknown hosts (optional) |
Syslog source |
Source address of syslog messages from unknown hosts (optional) |