Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials
Store

Interface Utilization for consecutive XX Minutes

Hi Forum,

Is there a way to report peak interface utilization which exceeds X% for more than XX consecutive minutes during business hours?

Thanks in advance.

Find more posts tagged with

interfaceutilization

Accepted answers

All comments

sunvad

@KMSigma.SWI @bobmarley - Can you help please?

bobmarley

I grabbed this query from a recent post. It's probably a good starting point. You could insert something on the Where clause tuned to your interface (or circuit) where H.InAveragebps is greater than x number. Same with H.OutAveragebps

SELECT tolocal (downsample  (h.DateTime ,'00:15:00')) as [Timestamp], 

N.Caption as [Device Name], I.Caption as [Interface Name], I.inBandwidth as [Bandwidth],
avg(H.InAveragebps) as [AverageIn],
avg(H.OutAveragebps) as [AverageOut]

FROM Orion.Nodes N
INNER JOIN Orion.NPM.Interfaces AS I 
ON N.NodeID = I.NodeID
INNER JOIN Orion.NPM.InterfaceTraffic AS H 
ON I.InterfaceID = H.InterfaceID 

--below will only show the last hour, adjust as necessary
WHERE N.Caption LIKE '%NODECAPTION%' AND I.Caption LIKE '%CIRCUITNAME%' AND h.DateTime > ADDDATE('Hour', -1, GETDATE())

GROUP BY 
N.Caption, I.Caption, I.inBandwidth,  downsample(H.DateTime,'00:15:00')

Order By Timestamp DESC

sunvad

Thank you @bobmarley, and sorry for my delayed response. I have seen this thread. Quick check please - does the query capture results, when the interface was utilized say more than 90% for a continuous time period of XX minutes? Not really good with scripting, sorry about that.

bobmarley

I put some comments in the query below.

SELECT tolocal (downsample (h.DateTime ,'00:5:00')) as [Timestamp], this is the average for 5 minutes. If you change it don't forget to change it in both places it shows up.

The circuit I was using was 10GB or 10,000,000,000. 90 percent of that is
9,000,000,000

That is shown in the Where clause as H.InAveragebps >= '9000000000' also H.OutAveragebps >= '9000000000'

This line tells the query to get only the last hour, adjust to your needs ADDDATE('Hour', -1, GETDATE()

The query will only report results for time periods that exceed 90 percent for each 5 minute average.

--Changed this down to 5 minute intervals to get all polls, match your interface polling time
SELECT tolocal (downsample  (h.DateTime ,'00:5:00')) as [Timestamp], 

N.Caption as [Device Name], I.Caption as [Interface Name], I.inBandwidth as [Bandwidth],
avg(H.InAveragebps) as [AverageIn],
avg(H.OutAveragebps) as [AverageOut]

FROM Orion.Nodes N
INNER JOIN Orion.NPM.Interfaces AS I 
ON N.NodeID = I.NodeID
INNER JOIN Orion.NPM.InterfaceTraffic AS H 
ON I.InterfaceID = H.InterfaceID 


--Put your node and interface name here ADDDATE('Hour', -1 tells the query how far to go back. The circuit I used had a bandwidth of 10000000000, 90 percent of that is 
--9000000000 so this will display only the times when the circuit was over 90 percent.

WHERE N.Caption LIKE '%YOURNODENAME%' AND I.Caption LIKE '%YOURCIRCUITNAME%' AND h.DateTime > ADDDATE('Hour', -1, GETDATE()) AND H.InAveragebps >= '9000000000' OR N.Caption LIKE '%bdc-rtr-isp-1111-1-NEW%' AND I.Caption LIKE '% Circuit ID: L4YS750288%' AND h.DateTime > ADDDATE('Hour', -1, GETDATE()) AND H.OutAveragebps >= '9000000000'

--Changed this down to 5 minute intervals to get all polls, match your interface polling time
GROUP BY
N.Caption, I.Caption, I.inBandwidth,  downsample(H.DateTime,'00:5:00')

Order By Timestamp DESC