cancel
Showing results for 
Search instead for 
Did you mean: 
Create Post
Level 14

Possible bug in NTA 2020.2

We upgraded our SolarWinds apps (including NTA) from 2019.4 HF4 to 2020.2 a few days ago.  At first everything seemed OK, until we checked the NTA application.  That's when things turned weird...

We are monitoring 360 routers for flow monitoring and close to 400 interfaces.  After the 2020.2 upgrade completed, flow monitoring has apparently stopped for over 95% of our routers. The SW tech I spoke with said there is a SolarWinds KB that says the solution was to remove the affected interfaces from SolarWinds, add them back and then re-select them for flow monitoring.  I am definitely not fond of that idea, since we'd lose all historical data for those interfaces.   

Have any of you run into this problem with NTA as well?   If so, have you been able to fix the problem without having to delete nodes and/or interfaces?  Thanks in advance.. 😎

4 Replies
Level 8

NTA stopped for us as well after the upgrade. The SW technician was able to get NTA services back for our Cisco Wireless Controller but our HP A10504 and A5800AF-48G switches, not so much.

0 Kudos
Level 7

Had the same issue, support confirmed a bug in the migration script for my incident which resulted in an interface with an incorrect local probe interface index in the NetflowInterfaceSources table.  I had already removed the affected interface which got everything working again, but it just needs the correct ID set.  

Unless your issue is different, they should have an internal bug for this problem so should be able to get you running again without too much effort.  

My issue appears to be quite different... 

***Most interfaces:  Flows received date is from before the upgrade began

***Some interfaces: Flows started and stopped again -- without reason

***A few interfaces:  Flows showing with current date

The SW tech already tried to fix by deleting and re-adding the duplicate interface.  By the way, in our environment, we have almost 400 interfaces being monitored, so the impact for us is quite sizable.  Anyway, still no luck, so we are still working w/ SW tech on this issue.🤕

0 Kudos

Probably not the same issue but the tech found a problem with script used during the upgrade process. They were able to remote in and correct the issue. We had to send them the diagnostic files from the primary server and a 15 minute Wireshark capture which helped them isolate the issue.

0 Kudos