PADD Cpu load 845% vs Proxmox LXC cpu load less than 1%

[woody4165]

Hi

I was getting a lot of this messages in PiHole installed in a Proxmox LXC container.

Long-term load (15min avg) larger than number of processors: 2.0 > 2This may slow down DNS resolution and can cause bottlenecks.

So I decided to install PADD to check what was going on and I found that in PADD the cpu load was over 800%, and increasing, while in the Proxmox cpu load is less then 5%, always, most of the times less than 1%

While I'm writing the cpu load in PADD has reached 1350%

What can I check ?

Thanks

[yubiuser]

This is not a PADD issue. It only reports, what the FTL API provides.
It was always difficult to report the 'right' CPU load on LXC containers, but recently we had a fix that should have fixed it (#2647).

[DL6ER]

What version of Pi-hole FTL are you using? As @yubiuser said, there was a fix what might have improved things starting with FTL v6.3. If you still see this, then you may need to flip on some special LXC switch ... what is needed is this project: https://github.com/lxc/lxcfs. Seems it will be used by default with LXC 1.1 or newer.

If you still observe these extraoridnarly high numbers with a recent version of FTL, you could maybe post some information from the relevant files lxcfs should touch once from inside the container and once from the host system:

/proc/cpuinfo
/proc/uptime
/sys/devices/system/cpu/online

[yubiuser]

What version of Pi-hole FTL are you using?

Looking at the posted PADD screenshot 6.3.3.

[woody4165]

thanks @yubiuser and @DL6ER

I'm using PADD 4.1.0 and FTL 6.3.3