August 18, 20232 yr I've "isolated" and "pinned" one core of my 16 core intel CPU for pfSense, and it seems to be working flawlessly except for a perhaps cosmetic defect? Nothing else is isolated or pinned. The CPU usage on the pfSense dashboard is always in the range 1-16%, but the CPU usage for that one core running the VM on the unraid dashboard is usually in the range 10-60%, sometimes peaking to 100% for a few seconds. I'm wondering why the big discrepancy? I understand they're measuring different things in different ways, but why to that large an effect? The highest usage (100%) seems to coincide with my GPU doing transcodes for Jellyfin or tdarr, but not always -- and neither of those are using much internet bandwidth so that seems weird too. Any ideas? Should I even care?
January 9, 20242 yr Author Nope. I just gave it the other core of the pair so it doesn't look "starved" on the usage graph, With two cores it barely ever gets above 5% usage. It's a waste, but makes me feel less angst.
October 6, 20241 yr Author On 1/9/2024 at 12:30 AM, Glide said: Nope. I just gave it the other core of the pair so it doesn't look "starved" on the usage graph, With two cores it barely ever gets above 5% usage. It's a waste, but makes me feel less angst. I think I found a work-around... I "isolated" one primary core and "pinned" it and the corresponding secondary core to the pfSense VM. Now the "usage" on both dashboards seems very similar.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.