[AusNOG] Switch Port Utilisation Monitoring

[Philip Loenneker]

Also, if a port is plugged into a server or router for example, it may never change state because you tend to have them on 24/7 ☺

From: AusNOG [mailto:ausnog-bounces at lists.ausnog.net] On Behalf Of Paul Wilkins
Sent: Thursday, 14 April 2016 3:02 PM
To: Nathan Phelan <nathan at interconnekt.com.au>
Cc: <ausnog at lists.ausnog.net> <ausnog at lists.ausnog.net>
Subject: Re: [AusNOG] Switch Port Utilisation Monitoring

The problem with traps, is you may have ports you never hear from. eg. If you have a mix of 24 and 48 ports, and you haven't heard from the top half of a switch. Essentially you'd have to create your own inventory of switchport interfaces.

Kind regards
Paul Wilkins

On 14 April 2016 at 14:31, Nathan Phelan <nathan at interconnekt.com.au<mailto:nathan at interconnekt.com.au>> wrote:
Logging SNMP traps to some sort of database (as per BB’s suggestion) sounds like the way to go.
If you don’t feel like rolling your own though you could try http://www.librenms.org/ (it’s a fork of Observium) – it has a per device event log (stored in mysql) which is searchable.
Cheers,
Nathan

From: AusNOG [mailto:ausnog-bounces at lists.ausnog.net<mailto:ausnog-bounces at lists.ausnog.net>] On Behalf Of Ben Buxton
Sent: Thursday, 14 April 2016 1:39 PM
To: Tim Raphael <raphael.timothy at gmail.com<mailto:raphael.timothy at gmail.com>>; Paul Wilkins <paulwilkins369 at gmail.com<mailto:paulwilkins369 at gmail.com>>
Cc: <ausnog at lists.ausnog.net<mailto:ausnog at lists.ausnog.net>> <ausnog at lists.ausnog.net<mailto:ausnog at lists.ausnog.net>>
Subject: Re: [AusNOG] Switch Port Utilisation Monitoring


Define "used".

Link up or actually carrying (meaningful) traffic?

For the former, you'd want to capture events as they happen, typically via snmp traps. Then just look at  how many reported a link up trap (including those that havent reported link down since the last interval).

For the latter, a raw packet count might suffice, but you'll have difficulty differentiating idle/keepalive traffic if the amount is low.

Personally, I'd go with snmp traps/polls injected into Prometheus, but many people havent the time or inclination to change their mindset into vector/timeseries based systems (despite the advantages).

BB

On Thu, Apr 14, 2016 at 1:28 PM Tim Raphael <raphael.timothy at gmail.com<mailto:raphael.timothy at gmail.com>> wrote:
Lets assume I have SNMP available.

Will Cacti be able to tell me over (for example) a week how many and which switch ports are used?

- Tim




On 14 Apr 2016, at 11:27 AM, Paul Wilkins <paulwilkins369 at gmail.com<mailto:paulwilkins369 at gmail.com>> wrote:

MRTG and/or Cacti.
(Assuming a "managed" switch ie. SNMP)

Kind regards
Paul Wilkins

On 14 April 2016 at 13:20, Tim Raphael <raphael.timothy at gmail.com<mailto:raphael.timothy at gmail.com>> wrote:
Hi All,

I would like a tool that could tell me exactly how many switch ports are in use across a given time period.
This is an enterprise environment so it’s not just the case of taking a snapshot in time, devices turn on and off so I’d want something that could monitor the port counts for a week or so and spit out a report.

Also, as per usual, the cheaper (free?) the better!

Any ideas?

- Tim


_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net<mailto:AusNOG at lists.ausnog.net>
http://lists.ausnog.net/mailman/listinfo/ausnog


_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net<mailto:AusNOG at lists.ausnog.net>
http://lists.ausnog.net/mailman/listinfo/ausnog

_______________________________________________
AusNOG mailing list
AusNOG at lists.ausnog.net<mailto:AusNOG at lists.ausnog.net>
http://lists.ausnog.net/mailman/listinfo/ausnog

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20160414/d0da97d3/attachment.html>