Network-discovery IPs sent with mask /32

[sepetuks]

I have following discovery configuration policy in agent config:

network_discovery:
  policy_1:
    config:
      defaults:
        tenant: Customer 1
        vrf: VRF1
      schedule: "*/5 * * * *"
      timeout: 5
    scope:
      targets:
         - 10.10.10.0/24
         - 10.10.10.254

• But in Diode server I see data load is with subnet /32 for all IP addresses (example: 10.10.10.4/32) that it detects. I think it is wrong as in IPAM model IP address should have subnet mask same as prefix and at least should follow definition in config.

• When only one IP address specified (second target) it is not possible to define subnet mask for it. If I set subnet mask it then scans full subnet but not single IP. As mentioned earlier it still loads all IPs with /32 mask,

[leoparente]

hi @sepetuks can you test the latest develop image by docker pull netboxlabs/orb-agent:develop. The expected version for network_discovery on it is 1.4.1.

This version has the capability to ** match target mask** by default.

And also in config:defaults: section you are able to overload the default /32 with any mask you want by defining network mask:

  policy_1:
    config:
      defaults:
        tenant: Customer 1
        vrf: VRF1
        network_mask: 24

you can check all the available options in: https://github.com/netboxlabs/orb-agent/blob/develop/docs/backends/network_discovery.md

[sepetuks]

Hi,
It ingested with mask /24 based on scope. (that's good)

But I tried to use override and looks override with "network_mask:24" doesn't work.

I tested with

 policy_1:
    config:
      defaults:
        tenant: Customer 1
        vrf: VRF1
        network_mask: 24
 scope:
      targets:
         - 10.10.10.0**/25**

in this case it was still ingesting 10.10.10.2/25 (so override didn't work)

[leoparente]

@sepetuks network_mask overrides the default /32. so if you have like a ip range 10.10.10.1-100, instead of having each ip with /32, It will apply the mask /24.

Now, regarding the issue that you had is because the new feature use_target_masks is enabled by default. To have the expected behavior of applying /24 in your case you need to disable it:

 policy_1:
    config:
      defaults:
        tenant: Customer 1
        vrf: VRF1
        network_mask: 24
 scope:
      targets:
         - 10.10.10.0/25
      use_target_masks: False

[sepetuks]

thanks. now it is clear.

[leoparente]

cool, I will close this ticket then.