Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Daniel Jakots
On Thu, 14 Apr 2022 23:47:42 +0200, Stefan Sperling wrote: > > $ sysctl kern.timecounter > > kern.timecounter.tick=1 > > kern.timecounter.timestepwarnings=0 > > kern.timecounter.hardware=pvclock0 > > kern.timecounter.choice=i8254(0) pvclock0(1500) acpihpet0(1000) > > acpitimer0(1000) > > > >

Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Tom Smyth
apologies all I missed (speed read Stuarts) mail... I would have a look at the preemption timer for the Host ... check out the top of page 15 of this amd manual... http://developer.amd.com/wp-content/resources/56263-Performance-Tuning-Guidelines-PUB.pdf I would try the two settings related to the

Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Tom Smyth
I have an Intel based Proxmox 7.1 being built pre-Production Ill have a go with it... Tomorrow and let you know On Thu, 14 Apr 2022 at 22:54, Tom Smyth wrote: > > Stuart, > sorry I wasnt entirely clear in my last email > > 1) you can try the /sys/module/kvm_intel/parameters/preemption_timer > >

Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Tom Smyth
Stuart, sorry I wasnt entirely clear in my last email 1) you can try the /sys/module/kvm_intel/parameters/preemption_timer if the system is an intel CPU based Physcial server 2) if you have an amd System you may find the issue does not occur in that case 3) looking at the DMESG I see a KVM CPU

Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Stefan Sperling
On Thu, Apr 14, 2022 at 09:26:41PM -, Stuart Henderson wrote: > I have some OpenBSD guests in Proxmox VE 7.1-7 (pve-qemu-kvm_6.1.0) and > seeing pretty bad clock drift (50 seconds in ~7h uptime). ntpd can't cope > with it. From boot: > > 2022-04-14T13:58:19.844Z ntpd[26996]: adjusting local

Re: time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Tom Smyth
Stuart is your host on an Intel System ? I had an awful time with Proxmox 5.0 and 5.1 with clock drift and console freezes can you try to disable the following feature in the Proxmox Host kernel /sys/module/kvm_intel/parameters/preemption_timer

time drift in OpenBSD in proxmox (qemu-kvm) guest

2022-04-14 Thread Stuart Henderson
I have some OpenBSD guests in Proxmox VE 7.1-7 (pve-qemu-kvm_6.1.0) and seeing pretty bad clock drift (50 seconds in ~7h uptime). ntpd can't cope with it. From boot: 2022-04-14T13:58:19.844Z ntpd[26996]: adjusting local clock by 1.745061s 2022-04-14T13:59:24.070Z ntpd[26996]: adjusting local

Re: Request additions to qbittorrent-nox port README

2022-04-14 Thread Stuart Henderson
+cc ports@ & maintainer On 2022/04/14 21:11, u...@mailo.com wrote: > > 127.0.0.1 is probably the best thing to suggest > > for listening to localhost. > The thing is - I need it accessible from another machine over network. > With `localhost` it DOES work over network, > this is how I have used

Re: How do I report a kernel panic occuring on install media?

2022-04-14 Thread rtw0 dtw0
Thanks for the tip! So I’ll try another configuration with another install. But so far I haven’t found any documentation about how to power off the system - disabling acpi prevents system from shutting down. Cheers Le jeu. 14 avr. 2022 à 12:29, Stuart Henderson a écrit : > On 2022/04/14

Re: How do I report a kernel panic occuring on install media?

2022-04-14 Thread Stuart Henderson
On 2022/04/14 12:21, rtw0 dtw0 wrote: > Hi, > > To disable acpi permanently: > # config -ef /bsd > ukc > disable acpi > ukc > quit This is a REALLY BAD IDEA. >From my earlier mail: https://marc.info/?l=openbsd-misc=164983204029245=2 | (Note: acpi drivers are used for various machine

Re: How do I report a kernel panic occuring on install media?

2022-04-14 Thread rtw0 dtw0
Hi, To disable acpi permanently: # config -ef /bsd ukc > disable acpi ukc > quit One thing I haven’t been able to do yet is get shutdown(8) to work with acpi disabled, as it ends up rebooting automatically and -p or -h options do not respond as they normally do. acpi(4) explains that acpi is