Same IP for Multiple Entities

Same IP for Multiple Entities

Problem

You see multiple entities in your monitoring UI with the same IP address but different names.

This can happen when the device using that IP has been renamed, or the IP has moved to a different device. The snmp-base.yaml file retains entries for every combination of device_name__ip that has ever been discovered. This is common when an IP is used for redundancy such as with HSRP, switch stacks, or primary/secondary failovers.

When a discovery job runs, each time it finds an IP it tries to determine the device_name connected to that address. If there is no existing entry in the devices list matching that specific device_name__ip combination, a new entry is added with the discovered device_name. Previous entries using that IP are not automatically overwritten or removed.

When ktranslate polls, it only uses the IP address — so if the devices list contains multiple entries with the same IP, it will collect and send metrics as if they were separate entities. In reality, all data is coming from whatever device responded to requests to that IP during the current polling interval. The device_name is not updated during polling.

Solution

One-time change (hardware replacement or naming convention update):

Edit snmp-base.yaml and delete the entry with the old device name. The old entity will stop receiving new data and will eventually age out of the system — typically 8 days after data stops coming in.

Redundancy groups or floating IPs:

Any IP that is part of a redundancy group or may move between devices should be added to the ignore_list in snmp-base.yaml. These devices should be monitored individually using a unique IP address that reliably connects to one specific device.

If you want to confirm that a virtual IP is reachable, you can add it manually as a ping_only device.