This discussion has been locked. The information referenced herein may be inaccurate due to age, software updates, or external references.

You can no longer post new replies to this discussion. If you have a similar question you can start a new discussion in this forum.

NPM 12.2 only Switches, Routers and Firewalls slow page loads

oldefortran over 5 years ago

Since upgrading to NPM 12.2 page loads are very slow only for Switches, Routers and Firewalls often taking 1-3 minutes to load the page.

Wireless Controllers, DWDM, UPS, VMAN and SRM are all fast with no problems.

We have 5 dedicated fairly high-end Cisco UCS servers 2.6 GHz dual quad-core with HT, 1 main NPM server, 3 APE's, a dedicated MS-SQL 2016 Server with RAID10.

When viewing Switches, Routers and Firewalls, CPU peaks around 35% max the rest of the time it's well below 20%. Main NPM and MS-SQL are on the same physical network segment. Each server has 2x 10 Gb/s NIC for a 20 Gb team.

I have defragmented database indexes. I shut down Activity Tracing that someone forgot and left running, but that did not resolve the problem.

Top Replies

d09h over 5 years ago +1

Tried Hubble? https://support.solarwinds.com/Success_Center/Network_Performance_Monitor_(NPM)/Turn_on_and_use_Hubble _

0 d09h over 5 years ago

Tried Hubble?
https://support.solarwinds.com/Success_Center/Network_Performance_Monitor_(NPM)/Turn_on_and_use_Hubble_
Cancel
Vote Up +1 Vote Down

Cancel
0 oldefortran over 5 years ago in reply to d09h

Thanks, that was very helpful, it seems weird with all of the searches I have done I have never seen Hubble mentioned.
The main holdup seems to be in the database query in Orion.HardwareHealth.HardwareItemBase which takes about 30 seconds. When I run the Database Maintenance manually it just hangs when it tries to process the hardware health table. I actually ran database maintenance before the upgrade and it processed that table with no problem at all. I had already tried Reorganize of the indexes for the table. It would be nice if I could just truncate the table and repopulate it somehow.
Cancel
Vote Up 0 Vote Down

Cancel
0 oldefortran over 5 years ago

With ProcessExplorer from SysInternals Suite I searched for handles with swdebugMaintenance and found a Database Maintenance process that was still stuck in the background, don't know why I didn't find it with regular Task Manager.
I killed that process, then was able to run Database Maintenance manually and it finally made it past the constipated Hardware Health tables and defragmented the indexes. I rebooted all the servers for good measure since I already had everything shut down. After reboot, I re-enabled hardware sensors and am currently getting about 9 seconds to load all pages including the Switches, Routers and Firewalls.
Cancel
Vote Up 0 Vote Down

Cancel