Nokia 7360 ISAM FX R6.8 SNMP OIDs for XGS-PON ONTs and uplink ports

Checkmk Dev&Exchange
1

Hi everyone,

we are currently integrating a Nokia 7360 ISAM FX OLT into Checkmk using SNMPv2c.

Our environment:

- Chassis: NFXS-E

- Software: R6.8.01r

- NT board: FANT-H

- LT board: FWLT-C

- Access technology: XGS-PON

- Monitoring system: Checkmk 2.5

Basic SNMP monitoring is already working. We can currently monitor the line cards and XGS-PON ports using Nokia enterprise OIDs under .1.3.6.1.4.1.637.

We can also see basic ONT objects through IF-MIB:

- ONT objects use ifType 270

- Physical UNI objects use ifType 271

- ifAdminStatus and ifOperStatus are available

However, we are still missing the correct Nokia MIBs or OIDs for:

- ONT ID and serial number

- ONT operational status

- ONT and OLT optical levels

- ONT distance

- Last offline reason

- LOS and Dying Gasp

- ONT UNI port status

- NT Ethernet port nt-a:eth:1

- XFP uplink port ntio-1:xfp:1

- LAG and LACP status

- Uplink traffic, errors, and optical DDM values

The standard IF-MIB does not provide useful names or complete status information for the NT and XFP uplink ports.

A full SNMP walk of .1.3.6.1.4.1.637 also times out, so we would prefer to query only the required Nokia subtrees.

Does anyone have a Nokia MIB package or an OID reference for the 7360 ISAM FX running R6.8, especially for FWLT-C, FANT-H, XGS-PON ONTs, and uplink monitoring?

We would also be interested to know whether the uplink ports must be queried using a separate SNMP context such as ihub.

Any examples, MIB names, OIDs, or existing monitoring plugins would

Thanks in advance!

Best regards
Alexander

2

Interesting that you even managed to get as far as you got without timing out.

We decided way back to use SNMP for the controller ethernet / uplink ports only, nt-a:xfp:1 and so on, and for the linecards we hacked a quick’n dirty special agent that uses pexpect to connect to the CLI via SSH and then execute “show equipment ont status x-pon” and similar commands and parse the XML response we get back.

But even this way (as compared to SNMP) trying to monitor any interface statistics would lead to very long agent runtime, and in any case the traffic statistics you get this way weren’t really helpful as Nokia doesn’t expose simple “bit/s” measurements as any other piece of equipment does, at least not via CLI.