Chart issues in SolarWinds NTA
Below are the most common issues encountered on SolarWinds NTA charts.
If your devices are configured to export NetFlow on both ingress and egress interfaces, you might see duplicate traffic in your resources.
Duplicate flows can occur in the following cases:
- You have both ip flow ingress and ip flow egress applied for all interfaces on a device.
- You have set ip flow ingress on some interfaces and ip flow egress on other interfaces.
- On your serial interfaces with subinterfaces, you have NetFlow export enabled on both the physical and logical interfaces.
Resolving Duplicate Flows
- If your device configuration contains both
ip flow ingressand
ip flow egresscommands, make sure NetFlow is enabled only for ingress interfaces.
- If you have NetFlow enabled for both physical and logical subinterfaces, remove the NetFlow export commands from the physical serial interfaces and only have the subinterfaces enabled for the export.
Double rate in Top XX Endpoints and Top XX IPv4 Domains
The Top XX Endpoints and Top XX IPv4 Domains resources display double data by design. Each flow has two distinct endpoints. To display statistics for top endpoints, SolarWinds NTA disregards that one endpoint is the source and the other endpoint is the target of flows, and treats both as endpoints only. This effectively doubles the total amount of data displayed by the Top XX Endpoints resource.
Let us take two flows and look at what you see in most resources and in the Top XX Endpoints resource.
|Flow||Source IP||Destination IP||Protocol||Bytes Transferred|
|Total bytes transferred:||50+40=90|
Top XX Endpoints resource
If your widgets show the "no data" message, it can be caused by one of the following:
No data to be displayed
There are no data to be displayed for the current time and flow direction settings.
To resolve the issue, check the time settings for both the resource and the view.
Too long time period selected for the view
If SolarWinds NTA needs more than one hour to process data that you want to see in the resources, the query times out and the resources show the "no data" message.
To resolve the issue, define a shorter time period for both the view and the resource.
Unexpected spikes in CBQoS post-policy charts
If you remove a shaping policy from a class, post-policy charts with the chart type set to % of Class Utilization may display unexpected spikes.
This is normal behavior, because devices affected by the policy change temporarily report huge amounts of data, which is reflected by the post-policy spike.