Re: High load average inside idle FreeBSD guest running on bhyve
--On 02 October 2014 15:34 +0200 Michael Gmelin wrote: For your bare metal box, could you send the output of vmstat -i ? I had the same issue on older boxes and it went away after setting hint.hpet.0.legacy_route=1 hint.attimer.0.clock=0 hint.atrtc.0.clock=0 in /boot/loader.conf (see hpet(4)). Sure, it's: " vmstat -i interrupt total rate irq4: uart0 724 0 irq17: ehci0 ehci1+ 113314 0 irq18: ohci0 ohci1*2 0 irq19: ahci0 12913 0 irq256: hpet0:t0 2208985 12 irq257: hpet0:t1 705576 4 irq259: re0 164805 0 irq260: re1 5049 0 irq261: re2 5049 0 Total3216417 18 " Box is an Alix APU (AMD G-T40E based, dual core @ 1Ghz). -Karl ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to "freebsd-virtualization-unsubscr...@freebsd.org"
Re: High load average inside idle FreeBSD guest running on bhyve
On Thu, 02 Oct 2014 11:49:19 +0100 Karl Pielorz wrote: > > > --On 02 October 2014 10:06 +0200 Christian Laursen > wrote: > > > Hello > > > > I have a few bhyve instances running that all seem to have a higher > > load average than what would seem normal for a system that is idle. > > > > [snip] > > > > last pid: 6109; load averages: 0.41, 0.38, 0.39up > > 1+14:27:48 09:56:25 > > Hi, > > I think this has been covered on other lists - and is apparently > "normal" (i.e. cosmetic). > > I've installed a number of FBSD10 boxes recently, and they all seem > to idle at around that (even when doing absolutely nothing) - I did > post about it and was told it was just cosmetic... It certainly seems > to be cosmetic. > > If it's the same issue - I'd guess it's not bhyve related... > > " > last pid: 1294; load averages: 0.42, 0.32, 0.27 > up 1+20:44:55 11:47:16 > 18 processes: 1 running, 17 sleeping > CPU: 0.0% user, 0.0% nice, 0.0% system, 0.1% interrupt, 99.9% idle > " > > That's from a bare metal amd64 10.0-RELEASE-p9 box with no services / > users on it. > > Thinking about it - we had the same 'symptom' on a Xen based FBSD 10 > box as well... For your bare metal box, could you send the output of vmstat -i ? I had the same issue on older boxes and it went away after setting hint.hpet.0.legacy_route=1 hint.attimer.0.clock=0 hint.atrtc.0.clock=0 in /boot/loader.conf (see hpet(4)). > > -Karl > > ___ > freebsd-virtualization@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization > To unsubscribe, send any mail to > "freebsd-virtualization-unsubscr...@freebsd.org" -- Michael Gmelin ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to "freebsd-virtualization-unsubscr...@freebsd.org"
Re: High load average inside idle FreeBSD guest running on bhyve
--On 02 October 2014 10:06 +0200 Christian Laursen wrote: Hello I have a few bhyve instances running that all seem to have a higher load average than what would seem normal for a system that is idle. [snip] last pid: 6109; load averages: 0.41, 0.38, 0.39up 1+14:27:48 09:56:25 Hi, I think this has been covered on other lists - and is apparently "normal" (i.e. cosmetic). I've installed a number of FBSD10 boxes recently, and they all seem to idle at around that (even when doing absolutely nothing) - I did post about it and was told it was just cosmetic... It certainly seems to be cosmetic. If it's the same issue - I'd guess it's not bhyve related... " last pid: 1294; load averages: 0.42, 0.32, 0.27 up 1+20:44:55 11:47:16 18 processes: 1 running, 17 sleeping CPU: 0.0% user, 0.0% nice, 0.0% system, 0.1% interrupt, 99.9% idle " That's from a bare metal amd64 10.0-RELEASE-p9 box with no services / users on it. Thinking about it - we had the same 'symptom' on a Xen based FBSD 10 box as well... -Karl ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to "freebsd-virtualization-unsubscr...@freebsd.org"
High load average inside idle FreeBSD guest running on bhyve
Hello I have a few bhyve instances running that all seem to have a higher load average than what would seem normal for a system that is idle. The host is running FreeBSD 10.1-BETA3 #2 r272300. FreeBSD 10.1-BETA3 #2 r272300: Tue Sep 30 16:26:53 CEST 2014 r...@trill.borderworlds.dk:/usr/obj/usr/src/sys/GENERIC amd64 FreeBSD clang version 3.4.1 (tags/RELEASE_34/dot1-final 208032) 20140512 CPU: Intel(R) Core(TM) i7-2600 CPU @ 3.40GHz (3411.56-MHz K8-class CPU) Origin = "GenuineIntel" Id = 0x206a7 Family = 0x6 Model = 0x2a Stepping = 7 Features=0xbfebfbff Features2=0x1fbae3ff AMD Features=0x28100800 AMD Features2=0x1 VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID TSC: P-state invariant, performance statistics The guests are all running FreeBSD 10.0-RELEASE-p9 updated via freebsd-update. The head of the top output inside a guest: last pid: 6109; load averages: 0.41, 0.38, 0.39up 1+14:27:48 09:56:25 24 processes: 1 running, 23 sleeping CPU: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle Mem: 6392K Active, 437M Inact, 379M Wired, 420M Buf, 3120M Free Swap: 4096M Total, 4096M Free I would have expected a load average around 0.00. Am I doing something wrong or is there something else that could be causing this? The guests all seem responsive and otherwise work fine. Apart from this I haven't encountered any problems when working with bhyve. It's nice and easy to work with. Thanks a lot for bringing this to FreeBSD. -- Christian Laursen ___ freebsd-virtualization@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-virtualization To unsubscribe, send any mail to "freebsd-virtualization-unsubscr...@freebsd.org"