SupermicroX8dt Dual Neahlem: FIXED Bad Clock Source (PVE2 crashes, time jitter etc)

plewka

Member
Sep 28, 2009
49
1
8
We went into strong troble when installing PVE2.1 on a new machine.
As soon there was some load the machine got slow and finally frozen
after slowly getting killed.

BUG: soft lockup - CPU#0 stuck for 70s! [events/0:35]

I had a look at the clocksource the kernel was using since it reported clock jitter.
It seems I was not able to disable any dynamic clocking in BIOS.
The kernel was using PIT or TSC as far as I know these are the timebase counters inside
the CPUs (don't thing there are separate ones for any core!?).
As soon as the CPUs use different clocking these no longer offer the same timebase.
I decided to move to HPET which is the most recent hardware concept of getting a usefull
timebase avoiding timedrifts missed RTC-IRQs and so on.

I added the parameter to the grub-config. Since there were messages regarding vt-d/iommu,
too - I disabled it too, because I don't need pci/pcie-passthru at that server.

/etc/default/grub

GRUB_CMDLINE_LINUX_DEFAULT="quiet intel_iommu=off clocksource=hpet console=tty2 console=ttyS2,115200n8"

Now the machine is rockstable:))

PS: Don't know why this HPET is not default today, btw.
 
Last edited:

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get yours easily in our online shop.

Buy now!