--Apple-Mail=_6C10F667-0B73-4BC1-9294-B0153AEB53AF Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=utf-8
On 28 Mar 2017, at 14:54, Gianluca Cecchi <gianluca.cecchi@gmail.com> = wrote: =20 On Tue, Mar 28, 2017 at 1:50 PM, Michal Skrivanek = <michal.skrivanek@redhat.com <mailto:michal.skrivanek@redhat.com>> = wrote: =20
On 28 Mar 2017, at 13:41, Ji=C5=99=C3=AD Sl=C3=A9=C5=BEka = <jiri.slezka@slu.cz <mailto:jiri.slezka@slu.cz>> wrote:
On 03/28/2017 12:34 AM, Ji=C5=99=C3=AD Sl=C3=A9=C5=BEka wrote:
Hi all,
I have just upgraded ovirt manager from 4.1.0 to 4.1.1, then I have upgraded one host. After reboot and activation of this host some = vms started to migrate here.
Some time after migration CPU of this vms go to 100% and vms become unreachable.
There is /var/log/libvirt/qemu/hypnos.log (hypnos is one of this = vms) https://pastebin.com/U30xfBMJ <https://pastebin.com/U30xfBMJ>
Powering off and starting again solve this but I would like to have = a glue what is happening.
I suspect this line
2017-03-27T22:09:23.746710Z qemu-kvm: warning: TSC frequency = mismatch between VM (2600031 kHz) and host (2100024 kHz), and TSC scaling = unavailable
Original host uses "AMD Opteron(TM) Processor 6238" with "cpu MHz: 2600.113" and upgraded host uses "AMD Opteron(tm) Processor 6172" = with "cpu MHz: 2100.097".
It looks like it isn't problem with 4.1.1 at all.
There is just missing tsc_scale cpu flag on our older AMD CPUs...
= http://unix.stackexchange.com/questions/296636/100-cpu-utilisation-and-han= g-after-virsh-migrate = <http://unix.stackexchange.com/questions/296636/100-cpu-utilisation-and-ha= ng-after-virsh-migrate>
This is probably unsolvable hw problem, right? =20 And this is without host-model CPU type? That 6172 should be Opteron G3=E2=80=A6is that what you have set in = your Cluster? =20 Thanks, michal =20 =20 =20 Should I care about it too? In my case on my AMD hypervisors: =20 [g.cecchi@ovmsrv06 ~]$ cat /proc/cpuinfo | grep tsc_ [g.cecchi@ovmsrv06 ~]$ =20 model name : Dual-Core AMD Opteron(tm) Processor 8222 flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge = mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext = fxsr_opt rdtscp lm 3dnowext 3dnow art rep_good nopl extd_apicid pni cx16 = lahf_lm cmp_legacy svm extapic cr8_legacy =20 The related cluster is set as "AMD Opteron G2=E2=80=9D
I suppose not unless you have newer Opterons. And if you do it=E2=80=99s = a matter of testing it I guess, whether the enforced =E2=80=9CAMD = Opteron G2=E2=80=9D on a newer Opteron works correctly.=20 Thanks, michal
=20 Thanks in advance, Gianluca
--Apple-Mail=_6C10F667-0B73-4BC1-9294-B0153AEB53AF Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=utf-8 <html><head><meta http-equiv=3D"Content-Type" content=3D"text/html = charset=3Dutf-8"></head><body style=3D"word-wrap: break-word; = -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" = class=3D""><br class=3D""><div><blockquote type=3D"cite" class=3D""><div = class=3D"">On 28 Mar 2017, at 14:54, Gianluca Cecchi <<a = href=3D"mailto:gianluca.cecchi@gmail.com" = class=3D"">gianluca.cecchi@gmail.com</a>> wrote:</div><br = class=3D"Apple-interchange-newline"><div class=3D""><div dir=3D"ltr" = class=3D""><div class=3D"gmail_extra"><div class=3D"gmail_quote">On Tue, = Mar 28, 2017 at 1:50 PM, Michal Skrivanek <span dir=3D"ltr" = class=3D""><<a href=3D"mailto:michal.skrivanek@redhat.com" = target=3D"_blank" class=3D"">michal.skrivanek@redhat.com</a>></span> = wrote:<br class=3D""><blockquote class=3D"gmail_quote" style=3D"margin:0px= 0px 0px 0.8ex;border-left:1px solid = rgb(204,204,204);padding-left:1ex"><span class=3D"gmail-"><br class=3D""> > On 28 Mar 2017, at 13:41, Ji=C5=99=C3=AD Sl=C3=A9=C5=BEka <<a = href=3D"mailto:jiri.slezka@slu.cz" class=3D"">jiri.slezka@slu.cz</a>> = wrote:<br class=3D""> ><br class=3D""> > On 03/28/2017 12:34 AM, Ji=C5=99=C3=AD Sl=C3=A9=C5=BEka wrote:<br = class=3D""> >> Hi all,<br class=3D""> >><br class=3D""> >> I have just upgraded ovirt manager from 4.1.0 to 4.1.1, then I = have<br class=3D""> >> upgraded one host. After reboot and activation of this host = some vms<br class=3D""> >> started to migrate here.<br class=3D""> >><br class=3D""> >> Some time after migration CPU of this vms go to 100% and vms = become<br class=3D""> >> unreachable.<br class=3D""> >><br class=3D""> >> There is /var/log/libvirt/qemu/hypnos.<wbr class=3D"">log = (hypnos is one of this vms)<br class=3D""> >> <a href=3D"https://pastebin.com/U30xfBMJ" rel=3D"noreferrer" = target=3D"_blank" class=3D"">https://pastebin.com/U30xfBMJ</a><br = class=3D""> >><br class=3D""> >> Powering off and starting again solve this but I would like to = have a<br class=3D""> >> glue what is happening.<br class=3D""> >><br class=3D""> >> I suspect this line<br class=3D""> >><br class=3D""> >> 2017-03-27T22:09:23.746710Z qemu-kvm: warning: TSC frequency = mismatch<br class=3D""> >> between VM (2600031 kHz) and host (2100024 kHz), and TSC = scaling unavailable<br class=3D""> >><br class=3D""> >> Original host uses "AMD Opteron(TM) Processor 6238" with "cpu = MHz:<br class=3D""> >> 2600.113" and upgraded host uses "AMD Opteron(tm) Processor = 6172" with<br class=3D""> >> "cpu MHz: 2100.097".<br class=3D""> ><br class=3D""> > It looks like it isn't problem with 4.1.1 at all.<br class=3D""> ><br class=3D""> > There is just missing tsc_scale cpu flag on our older AMD = CPUs...<br class=3D""> ><br class=3D""> > <a = href=3D"http://unix.stackexchange.com/questions/296636/100-cpu-utilisation= -and-hang-after-virsh-migrate" rel=3D"noreferrer" target=3D"_blank" = class=3D"">http://unix.stackexchange.com/<wbr = class=3D"">questions/296636/100-cpu-<wbr = class=3D"">utilisation-and-hang-after-<wbr class=3D"">virsh-migrate</a><br= class=3D""> ><br class=3D""> > This is probably unsolvable hw problem, right?<br class=3D""> <br class=3D""> </span>And this is without host-model CPU type?<br class=3D""> That 6172 should be Opteron G3=E2=80=A6is that what you have set in your = Cluster?<br class=3D""> <br class=3D""> Thanks,<br class=3D""> michal<br class=3D""> <div class=3D"gmail-HOEnZb"><div class=3D"gmail-h5"><br class=3D""><br = class=3D""></div></div></blockquote><div class=3D""><br = class=3D""></div><div class=3D"">Should I care about it too?</div><div = class=3D"">In my case on my AMD hypervisors:</div></div><br = class=3D""></div><div class=3D"gmail_extra"><div = class=3D"gmail_extra">[g.cecchi@ovmsrv06 ~]$ cat /proc/cpuinfo | grep = tsc_</div><div class=3D"gmail_extra">[g.cecchi@ovmsrv06 = ~]$</div></div><div class=3D"gmail_extra"><br class=3D""></div><div = class=3D"gmail_extra"><div class=3D"gmail_extra">model name<span = class=3D"gmail-Apple-tab-span" style=3D"white-space:pre"> </span>: = Dual-Core AMD Opteron(tm) Processor 8222</div><div class=3D"">flags<span = class=3D"gmail-Apple-tab-span" style=3D"white-space:pre"> = </span>: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov = pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp = lm 3dnowext 3dnow art rep_good nopl extd_apicid pni cx16 lahf_lm = cmp_legacy svm extapic cr8_legacy<br class=3D""></div><div class=3D""><br = class=3D""></div><div class=3D"">The related cluster is set as "AMD = Opteron G2=E2=80=9D</div></div></div></div></blockquote><div><br = class=3D""></div>I suppose not unless you have newer Opterons. And if = you do it=E2=80=99s a matter of testing it I guess, whether the enforced = =E2=80=9CAMD Opteron G2=E2=80=9D on a newer Opteron works = correctly. </div><div><br = class=3D""></div><div>Thanks,</div><div>michal</div><div><br = class=3D""><blockquote type=3D"cite" class=3D""><div class=3D""><div = dir=3D"ltr" class=3D""><div class=3D"gmail_extra"><div class=3D""><br = class=3D""></div><div class=3D"">Thanks in advance,</div><div = class=3D"">Gianluca</div></div></div> </div></blockquote></div><br class=3D""></body></html>= --Apple-Mail=_6C10F667-0B73-4BC1-9294-B0153AEB53AF--