Hi currently i have a librenms instance on ubuntu-server it has about 600 computers added and it is slow the virtual machine where i have it has resources to spare 45 gb of ram and 35 cpu cores of which it is using 10 perform an iotop and I could see that there is a process called [kworker/u66:2+flush-8:0]
using 99% of IO> I think this could be the problem I’m thinking of changing the priority to a much lower one to see if it works has anyone else had this happen?
I’ve only ever had issues in the past with AWS instances starving disk IO so not an expert in this, but worth running through these to ensure you’ve configured what you can, and have scaled out appropriately:
Digging in to kworker threads is in the realms of Linux support rather than LibreNMS, but below is one of the easier methods to have a closer look at what might be going on. You can select things and drill down a bit more with arrow keys/enter:
I’ve got little idea what most of that means, but it might highlight something to get you googling and trawling other Linux forums in the right direction. For example I now know that generate_Ku call is SNMPv3 converting a passphrase in to a user key with the requested encryption algorithm.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.