SolarWinds Knowledge Base

This article pertains to Orion NTA 3.5 and 3.6.

Record detailed results as you perform these steps. Doing so will help expedite a resolution should you need to contact SolarWinds Support about your CBQoS issue.

To troubleshoot a CBQoS issue:

1. Use packet capture on the relevant interface of the Orion server to verify SNMP (port 161) communication with relevant device(s).

    

2. Use SolarWinds MIB Viewer to check the status of CBQoS on your device.

   Support is determined by cbQosConfigIndex="1.3.6.1.4.1.9.9.166.1.5.1.1.2"

   If you see any value in your MIB viewer for this OID then CBQoS data is being successfully pulled.

3. Use Administrative Tools > Services to verify these services are running:

   SolarWinds Alerting Engine
SolarWinds Information Service
SolarWinds NetFlow
SolarWinds Network Performance Monitor
SolarWinds Information Service v1.

   In particular, the SolarWinds Orion Module Engine service enables polling of CBQoS. The NTA SolarWinds NetFlow service takes the data the Orion poller obtains from the device through SNMP and writes it into the NetFlow database table; and if this service is not working, polled CBQoS data will sit in a queue and eventually get dropped.

4. Verify no relevant routers have nested CBQoS policies.

   CBQoS collection for all devices will fail if just one device has nested policies. Check the NTA log file (swdebug.log) for a message like this:

   2009-07-22 00:23:53,453 [8] ERROR SolarWinds.Orion.NetFlow.ScheduledTask - CBQoSCollectData threw an exception. System.Data.SqlClient.SqlException: Violation of PRIMARY KEY constraint 'PK_cbQoSStats'. Cannot insert duplicate key in object 'dbo.cbQoSStats'.

   The statement has been terminated.

   Your only workaround is to manually disable CBQoS support for the relevant node(s) in the NTA Node Management resource (under Admin) on the Orion web console. (Note: This works for NTA 3.6).

5. Since Orion NTA polls for CBQoS data using Orion templates, verify that the NetFlow template exists and is enabled by issuing these SQL commands against the Orion database:

   Does the template exist?

   SELECT * FROM NPM_POLLER_TEMPLATES WHERE TemplateName = 'NetFlow' AND RowStatus = 0

   Table name – NPM_POLLER_TEMPLATES
   Column – ID = “TemplateID” ( Should be the same as active template in NPM_DEVICES table)
   Column – TemplateName = “Netflow”
   Column – RowStatus = “0” zero inticates active

   Is the template enabled?

   SELECT * FROM NPM_DEVICES WHERE RowStatus=0 and Enabled='True' and DvcID='NetFlow.cbQos'

   Table name – NPM_Devices
   Column – DvcID = “Netflow.cbQos”
   Column – Enabled = “True”
   Column – RowStatus = “0”

6. If the Netflow template either does not exist or is not enabled, start the Configuration Wizard in the Orion program group.

7. Check Database only, accept your current database settings, and allow the software wizard to perform database repairs.

8. To confirm the success of the repairs, repeat the query:

   SELECT * FROM NPM_POLLER_TEMPLATES WHERE TemplateName = 'NetFlow' AND RowStatus = 0

   If the repair was successful, the query will return exactly 1 row.

9. Assuming the Netflow template exists and is active, to verify that a selected node is marked as supporting CBQoS, issue this command to the Orion database:

   SELECT * FROM NPM_Nodes WHERE ModuleType = 4 AND State = 8 AND RowStatus = 0 AND OrionNodeID = [NODEID_HERE]

   State = 8 indicates that CBQoS is actively supported. So you should see your node in the query results.

10. If your node is CBQoS capable but is marked as not capable in the Orion, update the Orion database as follows to reset the discovery of devices so that supported features are recorded:

    UPDATE dbo.[NPM_NODES] SET State = 0 WHERE ModuleType = 4 AND RowStatus = 0

11. Use Administrative Tools > Services to restart the Orion Module Engine service.

    Device rediscovery could take up to 10 minutes after this service is restarted.

12. After a sufficient interval, repeat the query:

    SELECT * FROM NPM_Nodes WHERE ModuleType = 4 AND RowStatus = 0 AND OrionNodeID = [NODEID_HERE]

    You should see results in which State=8.

13. Assuming the attempt to reset device discovery failed, contact SolarWinds Support with the results of your troubleshooting so far.