I already gave you the two OIDs in the previous mail.

The existence or not of openbsd specific MIBs are irrelevant in this instance, since its using IF-MIB (also, MIBs don't really do what you think they do)

Adam.

Sent from BlueMail

On 24 May 2016, at 18:14, Andy Lemin <andy@brandwatch.com> wrote:

Yea I figured the same which is why I was after the OID being called?

Does anyone know? Or know how to find out? Where to look?

OpenBSD actually runs its own SNMPD (not net-snmpd). In the past net-snmpd gave you all the generic stuff, but nothing from the OpenBSD enterprise. And SNMPD gave you everything from the OpenBSD enterprise, but not the general stuff.

So I have seen people running net-snmpd on the interfaces, and SNMPD only on the loopback, then setting up an OID forwarder within net-snmp to forward all OpenBSD enterprise OIDs to the localhost etc. This kinda worked but was fragmented and had issues and data holes.. :(

However a year ago or so, SNMPD was significantly extended so that it now provides everything net-snmpd does, and so we only run SNMPD. If SNMPD is returning the wrong value somewhere, then I can definitely take that to the OpenBSD guys.

But before I can I need to know the OID Observium is calling ;)

Also seeing as SNMPD was extended, I think making sure Observium has the latest OpenBSD MIB's could be important.

Thanks, Andy.

On Tue, May 24, 2016 at 2:33 PM, Adam Armstrong <adama@memetic.org> wrote:

There's virtually no chance of this bring anything other than net SNMP on openbsd rather helpfully reporting the same ifInDiscards ifOutDiscards counters for all interfaces.

What version of net-snmp is it?

Adam.

Sent from BlueMail

On 24 May 2016, at 14:31, Andy Lemin <andy@brandwatch.com> wrote:

Hi,

We have noticed an interesting problem where on all of our OpenBSD firewalls, every single interface on the same firewall reports the same discards rate.

We would like to try and work out which interface these discards are actually happening on, so we can work out what the cause is and check for a possible remedy (wrong VLANs being trunked through for example). If they are bad checksums, than that's ok. But would like to know ;)

Below is a screenshot showing the "Errors" graphs for all ports on the same server, notice how they are all displaying the same data. 

My thoughts of how to start are;

   What is the SNMP OID being polled for this information?

I wonder if OpenBSD is incorrectly reporting the same value on all interfaces, or if there is only a single Discards counter/OID for the whole system, or if it is an Observium problem etc..

So if anyone can tell me the OID, I can do some more investigation first ;)

I don't mind setting up OID mappings in SNMPD etc, if that is required I would be happy to provide a configuration template for Observium and OpenBSD.

And I also wonder how up to date the OpenBSD MIB file built into Observium is?

Thanks in advance for your help :)

Kind regards, Andy.

​

observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium