Welcome! Use the login menu to access:
Customer Portal
SolarWinds customers can download purchased software, renew maintenance, & more in our updated Customer Portal!
Community
After logging in, community members can rate & review SolarWinds software & participate in forums on thwack.com.
Partner Portal
SolarWinds channel partners will find all the resources they need in the Partner Portal, including exclusive partner tools.
| More

SolarWinds Knowledge Base

NetFlow Interface resource(s) show utilization peaks higher than NPM utilization or higher than the interface bandwidth

Resolution

Flow exporters may store active flows for longer than the default NTA reporting period of 1 minute. When long duration flows are exported the data is reported as occurring in same minute the export is sent to NTA. The result may cause reporting of data rates larger than the actual rate for that minute.

The active flow cache timer on each flow exporter should be set to 1 minute to match the NTA reporting period.

For Cisco devices running IOS 12.2 (14)S and above, include ip flow-cache active 1 in the device global configuration.

For Adtran devices running 16.1 and above, include ip flow cache timeout active 1 in the device configuration.

For sFlow: set polling interval 60 Exact syntax may vary by vendor implementation. See your vendor’s configuration guide for details.

For NetFlow (Catalyst switches): mls aging long 64 in the device global configuration. 64 seconds is the minimum allowed value.

J-Flow: ip flow-cache timeout active 1 in the device configuration.

 

Note:

There is a known issue in Orion NTA 3.6 (resolved in 3.7) that can rarely cause similar behavior. You can detect if this bug can cause this behavior by running the following query using MS SQL Management Studio when connected to your Orion database:

SELECT

      n1.caption,

      n2.caption,

      i.InterfaceName,

      a.InterfaceName

FROM

      interfaces i

      INNER JOIN

            (

            SELECT (nodeId * 4096) | interfaceIndex as hash, InterfaceIndex, InterfaceName, NodeId, interfaceid

            FROM interfaces

            ) A

            ON A.hash = ((i.nodeId * 4096) | i.interfaceIndex)

                  AND NOT(a.interfaceindex = i.interfaceindex AND a.nodeid = i.nodeid)

      INNER JOIN Nodes n1 on i.NodeID = n1.NodeID

      INNER JOIN Nodes n2 on a.NodeID = n2.NodeID

WHERE

      n1.NodeID < n2.NodeID

If the query returns results, then the results indicate nodes (and their interfaces) where you will see incorrect traffic results. In this case, please upgrade to 3.7.



Related Articles

Attachments

No attachments were found.

Visitor Comments

No visitor comments posted. Post a comment

Post Comment for "NetFlow Interface resource(s) show utilization peaks higher than NPM utilization or higher than the interface bandwidth"

If this article does not provide the solution you need, please consider visiting our community forums on thwack or contacting SolarWinds Support. Please use the following form to provide comments that help us improve this KB article. Fields marked with an asterisk are required.

   Name:
* Email:
* Comment:
* Enter the code below:

 

Article Details

Last Updated
2nd of September, 2010

Would you like to...

Print this page  Print this page

Email this page  Email this page

Post a comment  Post a comment

 Subscribe me

Subscribe me  Add to favorites

Remove Highlighting Remove Highlighting

Edit this Article

Quick Edit

Export to PDF


User Opinions

100% thumbs up
(1 vote)
0% thumbs down
(0 votes)

How would you rate this answer?




Thank you for rating this answer.

Continue