https://bugzilla.kernel.org/show_bug.cgi?id=61971 Bug ID: 61971 Summary: Migration problem of qemu-kvm guests Product: Virtualization Version: unspecified Kernel Version: 3.10.5 Hardware: x86-64 OS: Linux Tree: Mainline Status: NEW Severity: normal Priority: P1 Component: kvm Assignee: virtualization_kvm@xxxxxxxxxxxxxxxxxxxx Reporter: magnus@xxxxxxxx Regression: No Hello, I am having migration problems and would appreciate very much help in troubleshooting this further. I have installed two wheezy hosts (qemu-1.1.2) using glusterfs as backend storage and are running about 12 debian wheezy guests. Node A has newer hardware (AMD FX(tm)-8320 Eight-Core Processor) and node B is older (AMD Phenom(tm) II X6 1055T Processor). I am using libvirt and when migrating hosts from Node A to node B the always freeze indicating high cpu utilization in virt-manager for example. Migrations from Node B to Node A always works. I can start all gests on both hosts without problems. To rule out glusterfs as the problem I have also used nfs as backend storage with the same results. I have installed a freebsd-9.1 guest to rule out the guests linux kernel as the likely problem. The guests also freeze if they are in grub or the freebsd bootloader while being migrated. I have tried using this kernel parameter processor.max_cstate=1 and using changing clocksource to acpi_pm (defaults to kvm-clock). Neither of these had any effect on my problem. I have tried disabling the balloon driver and that didn't help either. Both hosts and all guests use ntp so the time is synced. At one point I reinstalled both machines as ubuntu 13.04 to get newer versions of qemu (1.4) and libvirt but the exact same problem was there also, as I prefer debian I went back to wheezy. -no-kvm-irqchip and -no-kvm-pit doesn't help either. Using -no-kvm the migration works both ways. My gut feeling is that it has something to do with the different cpu's, but I read that it was possible to migrate between intel and amd I guess this shouldn't be a problem. Anyone have any ide what could be the problem here? Best Regards Magnus Diff between cpuinfo of linux guest when booted on each host so no cpu flags differ in the guest. I have tried with Opteron_G2, qemu64 and kvm64 cpus all causes the guest to freeze. root@lazy:~# diff -Nru iron1-cpuinfo.txt iron2-cpuinfo.txt --- iron1-cpuinfo.txt 2013-09-19 20:59:24.407659093 +0200 +++ iron2-cpuinfo.txt 2013-09-19 21:01:03.326022357 +0200 @@ -5,14 +5,14 @@ model name : QEMU Virtual CPU version 1.1.2 stepping : 3 microcode : 0x1000065 -cpu MHz : 3511.722 +cpu MHz : 2812.460 cache size : 512 KB fpu : yes fpu_exception : yes cpuid level : 4 wp : yes flags : fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx lm up nopl pni cx16 popcnt hypervisor lahf_lm svm abm sse4a -bogomips : 7023.44 +bogomips : 5624.92 TLB size : 1024 4K pages clflush size : 64 cache_alignment : 64 Node A cpuinfo processor : 7 vendor_id : AuthenticAMD cpu family : 21 model : 2 model name : AMD FX(tm)-8320 Eight-Core Processor stepping : 0 microcode : 0x600081f cpu MHz : 1400.000 cache size : 2048 KB physical id : 0 siblings : 8 core id : 7 cpu cores : 4 apicid : 23 initial apicid : 7 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes 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 pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid_msr tbm topoext perfctr_core perfctr_nb arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold bmi1 bogomips : 7023.38 TLB size : 1536 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm 100mhzsteps hwpstate cpb eff_freq_ro Node B cpuinfo processor : 5 vendor_id : AuthenticAMD cpu family : 16 model : 10 model name : AMD Phenom(tm) II X6 1055T Processor stepping : 0 microcode : 0x10000dc cpu MHz : 800.000 cache size : 512 KB physical id : 0 siblings : 6 core id : 5 cpu cores : 6 apicid : 5 initial apicid : 5 fpu : yes fpu_exception : yes cpuid level : 6 wp : yes 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 pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt cpb hw_pstate npt lbrv svm_lock nrip_save pausefilter bogomips : 5624.50 TLB size : 1024 4K pages clflush size : 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm stc 100mhzsteps hwpstate cpb Libvirt starts the guests like this: LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin QEMU_AUDIO_DRV=none /usr/bin/kvm -S -M pc-1.1 -enable-kvm -m 512 -smp 2,sockets=2,cores=1,threads=1 -name jefster -uuid f94a85d4-629e-bdba-b585-282638526800 -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/jefster.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=utc -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/vmimages/gv0/jefster.img,if=none,id=drive-virtio-disk0,format=qcow2 -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -netdev tap,fd=26,id=hostnet0,vhost=on,vhostfd=29 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:61:9b:f3,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -device usb-tablet,id=input0 -vnc 127.0.0.1:5 -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 Hosts kernel: Linux iron1 3.10-0.bpo.2-amd64 #1 SMP Debian 3.10.5-1~bpo70+1 (2013-08-11) x86_64 GNU/Linux -- You are receiving this mail because: You are watching the assignee of the bug. -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
