[ovirt-devel] [vdsm] strange network test failure on FC23
Michal Skrivanek
mskrivan at redhat.com
Mon Nov 30 08:52:59 UTC 2015
> On 29 Nov 2015, at 17:34, Nir Soffer <nsoffer at redhat.com> wrote:
>
> On Sun, Nov 29, 2015 at 6:01 PM, Yaniv Kaul <ykaul at redhat.com> wrote:
> > On Sun, Nov 29, 2015 at 5:37 PM, Nir Soffer <nsoffer at redhat.com> wrote:
> >>
> >> On Sun, Nov 29, 2015 at 10:37 AM, Yaniv Kaul <ykaul at redhat.com> wrote:
> >> >
> >> > On Fri, Nov 27, 2015 at 6:55 PM, Francesco Romani <fromani at redhat.com>
> >> > wrote:
> >> >>
> >> >> Using taskset, the ip command now takes a little longer to complete.
I fail to find the original reference for this.
Why does it take longer? is it purely the additional taskset executable invocation? On busy system we do have these issues all the time, with lvm, etc…so I don’t think it’s significant
> >> >
> >> >
> >> > Since we always use the same set of CPUs, I assume using a mask (for 0 &
> >> > 1,
> >> > just use 0x3, as the man suggests) might be a tiny of a fraction faster
> >> > to
> >> > execute taskset with, instead of the need to translate the numeric CPU
> >> > list.
> >>
> >> Creating the string "0-<last cpu index>" is one line in vdsm. The code
> >> handling this in
> >> taskset is written in C, so the parsing time is practically zero. Even
> >> if it was non-zero,
> >> this code run once when we run a child process, so the cost is
> >> insignificant.
> >
> >
> > I think it's easier to just to have it as a mask in a config item somewhere,
> > without need to create it or parse it anywhere.
> > For us and for the user.
>
> We have this option in /etc/vdsm/vdsm.conf:
>
> # Comma separated whitelist of CPU cores on which VDSM is allowed to
> # run. The default is "", meaning VDSM can be scheduled by the OS to
> # run on any core. Valid examples: "1", "0,1", "0,2,3"
> # cpu_affinity = 1
>
> I think this is the easiest option for users.
+1
>
> >> > However, the real concern is making sure CPUs 0 & 1 are not really too
> >> > busy
> >> > with stuff (including interrupt handling, etc.)
> >>
> >> This code is used when we run a child process, to allow the child
> >> process to run on
> >> all cpus (in this case, cpu 0 and cpu 1). So I think there is no concern
> >> here.
> >>
> >> Vdsm itself is running by default on cpu 1, which should be less busy
> >> then cpu 0.
> >
> >
> > I assume those are cores, which probably in a multi-socket will be in the
> > first socket only.
> > There's a good chance that the FC and or network/cards will also bind their
> > interrupts to core0 & core 1 (check /proc/interrupts) on the same socket.
> > From my poor laptop (1s, 4c):
> > 42: 1487104 9329 4042 3598 IR-PCI-MSI 512000-edge
> > 0000:00:1f.2
> >
> > (my SATA controller)
> >
> > 43: 14664923 34 18 13 IR-PCI-MSI 327680-edge
> > xhci_hcd
> > (my dock station connector)
> >
> > 45: 6754579 4437 2501 2419 IR-PCI-MSI 32768-edge
> > i915
> > (GPU)
> >
> > 47: 187409 11627 1235 1259 IR-PCI-MSI 2097152-edge
> > iwlwifi
> > (NIC, wifi)
>
> Interesting, here an example from a 8 cores machine running my vms:
>
> [nsoffer at jumbo ~]$ cat /proc/interrupts
> CPU0 CPU1 CPU2 CPU3 CPU4 CPU5 CPU6 CPU7
> 0: 31 0 0 0 0 0 0 0 IR-IO-APIC-edge timer
> 1: 2 0 0 1 0 0 0 0 IR-IO-APIC-edge i8042
> 8: 0 0 0 0 0 0 0 1 IR-IO-APIC-edge rtc0
> 9: 0 0 0 0 0 0 0 0 IR-IO-APIC-fasteoi acpi
> 12: 3 0 0 0 0 0 1 0 IR-IO-APIC-edge i8042
> 16: 4 4 9 0 9 1 1 3 IR-IO-APIC 16-fasteoi ehci_hcd:usb3
> 23: 13 1 5 0 12 1 1 0 IR-IO-APIC 23-fasteoi ehci_hcd:usb4
> 24: 0 0 0 0 0 0 0 0 DMAR_MSI-edge dmar0
> 25: 0 0 0 0 0 0 0 0 DMAR_MSI-edge dmar1
> 26: 3670 354 215 9062370 491 124 169 54 IR-PCI-MSI-edge 0000:00:1f.2
> 27: 0 0 0 0 0 0 0 0 IR-PCI-MSI-edge xhci_hcd
> 28: 166285414 0 3 0 4 0 0 0 IR-PCI-MSI-edge em1
> 29: 18 0 0 0 4 3 0 0 IR-PCI-MSI-edge mei_me
> 30: 1 151 17 0 3 169 26 94 IR-PCI-MSI-edge snd_hda_intel
> NMI: 2508 2296 2317 2356 867 918 912 903 Non-maskable interrupts
> LOC: 302996116 312923350 312295375 312089303 86282447 94046427 90847792 91761277 Local timer interrupts
> SPU: 0 0 0 0 0 0 0 0 Spurious interrupts
> PMI: 2508 2296 2317 2356 867 918 912 903 Performance monitoring interrupts
> IWI: 1 0 0 5 0 0 0 0 IRQ work interrupts
> RTR: 0 0 0 0 0 0 0 0 APIC ICR read retries
> RES: 34480637 12953645 13139863 14309885 8881861 10110753 9709070 9703933 Rescheduling interrupts
> CAL: 7387779 7682087 7283716 7135792 2771105 1785528 1887493 1843734 Function call interrupts
> TLB: 11121 16458 17923 15216 8534 8173 8639 7837 TLB shootdowns
> TRM: 0 0 0 0 0 0 0 0 Thermal event interrupts
> THR: 0 0 0 0 0 0 0 0 Threshold APIC interrupts
> MCE: 0 0 0 0 0 0 0 0 Machine check exceptions
> MCP: 7789 7789 7789 7789 7789 7789 7789 7789 Machine check polls
> HYP: 0 0 0 0 0 0 0 0 Hypervisor callback interrupts
> ERR: 0
> MIS: 0
>
> It seems that our default (CPU1) is fine.
I think it’s safe enough.
Numbers above (and I checked the same on ppc with similar pattern) are for a reasonablt epty system. We can get a different picture when vdsm is busy. In general I think it’s indeed best to use the second online CPU for vdsm and all CPUs for child processes
regarding exposing to users in UI - I think that’s way too low level. vdsm.conf is good enough
Thanks,
michal
>
> Francesco, what do you think?
>
> Nir
More information about the Infra
mailing list