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 Bandwidth" sometimes reports incorrect size??

Sometimes, the "Interface Bandwidth" in the Interface Details page will show up with a size of 10.0 Mbps, when it should be reporting 100 or 1000 Mbps. Anyone know why this happens?? My pollers are well within load limits.

I am running NPM 9.5.1 on all pollers. Ticket 146076 just opened regarding this. This problem occurs intermittently, not always on the same node, and not always on the same interface. I am uncertain if it always occurs on nodes being monitored by the same polling engine.

I just purchased and brought a 4th polling engine online (upgrading the primary and 2 secondaries from 9.5.0 to 9.5.1 at the same time). I have ~18,000 total elements monitored with more being added, mainly spread across the 3 secondary pollers, with the primary poller having next to no load, as I plan to turn up NTA again soon. I noticed this "Interface Bandwidth" issue both before and after bringing the 4th poller online to reduce load.

I would appreciate suggestions, answers, help.

Find more posts tagged with

Accepted answers

warbird

In my case, 3 things lead to this problem. 2 of which I was able to address, 1 of which is not addressable. The first 2 were load issues and db issues. My polling engines were being taxed beyond their limits and I had very large tables with very old data in my db that were not rolling up properly (we are currently monitoring ~1300 devices and ~23K elements). This was evidenced by errors in the db maintenance logs. SW support helped me address the db issues and I brought another poller online to address the load issues.

I am now running v10 SP1 and I still occasionally notice a few interfaces reporting incorrect "Interface Bandwidth" from time to time. As best I can tell, NPM is reporting accurately. Here is why I think this...

Certain Cisco switches with ports set to auto/auto will report to NPM as 10M when they are enabled but not active. If that port goes active, negotiates at 1G, and the client on that port starts pushing a bunch of data, Orion will display what you are describing until the next full statistics poll of that device.

Is there a way for SW to fix this by moving the "Interface Bandwidth" poll such that it happens at the same time as the Interface Usage poll? I do not know but I imagine that would heavily increase the overall polling load for something that doesn't occur often if systems are running as they should.

As previously stated, I recommend you open a SW support ticket. Have them take a look at system loads and db logs with you. Perhaps that will help you. I cannot speak about manually running the discovery tool, as I don't use it.

HTH. As always, YMMV.

All comments

warbird

I should add that when this problem occurs, it skews my graphs and messes up my "Top 10" type of reports, wrongly showing an interface as flooded. This is a huge problem for us.

sraiwt

This happens to us all the time as well. It's quite annoying and would like it fixed.

warbird

Quick update: I am working with support to track down the issue. I have cleaned up a few issues with the db (possibly unrelated), and they are evaluating a new set of diagnostics.

sraiwt, how many polling engines are you running? What is your environment like? It seems like the problem is a bit more noticeable since I brought on the 4th polling engine. Not positive, though.

I will keep you posted here on our progress.

sraiwt

We have a rather small environment compared to alot of the users I see on here. We're currently using one single box for everything (DB, Poller, Web) and we're running APM and the wireless plugins. We have approx 11k elements as well in our environment.

warbird

In my experience, that is a lot to have on a single box. Glad it is working for you. You must have polling intervals turned down? Our polling engines start choking around 7k elements. I also have our db on a standalone server with the actual db backended on an extremely beefy SAN (we had run into issues with disk queue length problems).

The issue reported in this thread is still occurring this morning. So the db cleanup I did yesterday did not work.