topic Re: SNMP Monitoring on port Utilization in Switches, Hubs, and Modems

SNMP Monitoring on port Utilization

joe_butler — Thu, 24 Sep 2009 14:30:21 GMT

Hi Everyone,

I am trying to monitor port utilization on my trunks, using SNMP to alert me if they stay maxed for any length of time. I have done a mib walk on one of the switches but nothing seems to match up with the current speed however all the other interface stats are included.

Has anyone managed to monitor current util on ports using SNMP?

Re: SNMP Monitoring on port Utilization

Steve Britt — Thu, 24 Sep 2009 15:33:27 GMT

Hi Joe,

I don't know if you've tried out ProCurve Manager yet, but it has a traffic monitoring capability that is capable of what you're asking. It can collect traffic either through SNMP polling or via sFlow - you can control which one on each port or let it decide for you - and it reports several key stats for each port/trunk/mesh monitored every minute (including utilization). Trunk and mesh stats are reported for the aggregate virtualized interface rather than the separate individual ports that make up the trunk/mesh. You can set ProCurve Manager up so that it generates an alert to you when specific thresholds are crossed (e.g. critical alert when utilization exceeds 95%, warning alert when it exceeds 85%, etc.).

You can try ProCurve Manager for free for 60 days to see if it meets your needs. It can be downloaded from the ProCurve web site.

Regards,

SVB

Re: SNMP Monitoring on port Utilization

joe_butler — Thu, 24 Sep 2009 15:44:54 GMT

Hi Steve,

I do have PCM and it can pull back trunk utilisation. However i think its a very poor piece of software. I'm using PRTG for network monitoring which I was hoping use the oid in to regularly pull back the current traffic on the trunk.

sFlow would be an option, I have written some custom tools to collect sampling data from my procurves, I havent looked too far into it but I think that using sFlow polling on the trunk ports might to the trick!

I would still like to know if there is a specific OID to measure port util tho, with this I could write a script to run on demand, and gather the current status on the trunks