Description of problem: Running a custom guest OS (based on Fedora Core 12), the guest crash with a kernel panic after a few days : [ 134.251405] BUG: unable to handle kernel paging request at ffffffff8104e8f7 [ 134.252284] IP: [<ffffffff8104e8f7>] kvm_kick_cpu+0x27/0x30 [ 134.252979] PGD 1c0a067 [ 134.253266] PUD 1c0b063 PMD 10001e1 [ 134.253733] [ 134.253936] Oops: 0003 [#1] SMP [ 134.254316] Modules linked in: softdog virtio_net cpuid tg3 hwmon e1000e ptp pps_core e1000 vmxnet3 i2c_i801 i2c_smbus virtio_console xt_conntrack ipt_MASQUERADE nf_nat_masquerade_ipv4 nf_conntrack_netlink xt_addrtype iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack overlay [last unloaded: virtio_net] [ 134.258187] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 4.9.58 #1 [ 134.259039] Hardware name: KubeVirt None, BIOS 1.12.0-5.module+el8.1.1+5309+6d656f05 04/01/2014 [ 134.259256] task: ffffffff81c0e500 task.stack: ffffffff81c00000 [ 134.259256] RIP: 0010:[<ffffffff8104e8f7>] [<ffffffff8104e8f7>] kvm_kick_cpu+0x27/0x30 [ 134.259256] RSP: 0018:ffff88107fc03df0 EFLAGS: 00010046 [ 134.259256] RAX: 0000000000000005 RBX: 0000000000000000 RCX: 0000000000000002 [ 134.259256] RDX: ffff88107fc80000 RSI: 0000000000000003 RDI: 0000000000000002 [ 134.259256] RBP: ffff88107fc03df8 R08: 0000000000000068 R09: ffff88107ffe5640 [ 134.259256] R10: 0000000000000100 R11: ffffea004093f9c0 R12: ffff881036fa0f00 [ 134.259256] R13: 0000000000000001 R14: ffff881038a09000 R15: 0000000000000082 [ 134.259256] FS: 0000000000000000(0000) GS:ffff88107fc00000(0000) knlGS:0000000000000000 [ 134.259256] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 134.259256] CR2: ffffffff8104e8f7 CR3: 0000001038131000 CR4: 00000000001406b0 [ 134.259256] Stack: [ 134.259256] ffff881036fa4800 ffff88107fc03e08 ffffffff810a2c90 ffff88107fc03e58 [ 134.259256] ffffffff810a2665 ffffea004093f9c0 ffff881036818350 0000000000000000 [ 134.259256] ffff881036818340 ffff881036fa4808 0000000000000003 0000000000000000 [ 134.259256] Call Trace: [ 134.259256] <IRQ> [ 134.259256] [<ffffffff810a2c90>] __pv_queued_spin_unlock_slowpath+0xa0/0xe0 [ 134.259256] [<ffffffff810a2665>] __raw_callee_save___pv_queued_spin_unlock_slowpath+0x15/0x24 [ 134.259256] [<ffffffff810a2694>] .slowpath+0x9/0x15 [ 134.259256] [<ffffffff81758aa0>] _raw_spin_unlock_irqrestore+0x10/0x20 [ 134.259256] [<ffffffff81415ad8>] virtblk_done+0xc8/0xe0 [ 134.259256] [<ffffffff8139a301>] vring_interrupt+0x31/0x50 [ 134.259256] [<ffffffff810aa4c1>] __handle_irq_event_percpu+0x41/0x1b0 [ 134.259256] [<ffffffff810aa653>] handle_irq_event_percpu+0x23/0x60 [ 134.259256] [<ffffffff810aa6cb>] handle_irq_event+0x3b/0x60 [ 134.259256] [<ffffffff810ad98b>] handle_edge_irq+0xab/0x150 [ 134.259256] [<ffffffff8102cadd>] handle_irq+0x1d/0x30 [ 134.259256] [<ffffffff8175b35d>] do_IRQ+0x4d/0xd0 [ 134.259256] [<ffffffff817598bf>] common_interrupt+0x7f/0x7f [ 134.259256] <EOI> [ 134.259256] [<ffffffff81758816>] ? native_safe_halt+0x6/0x10 [ 134.259256] [<ffffffff81758563>] default_idle+0x23/0xd0 [ 134.259256] [<ffffffff810341ff>] arch_cpu_idle+0xf/0x20 [ 134.259256] [<ffffffff81758973>] default_idle_call+0x23/0x40 [ 134.259256] [<ffffffff810a05bc>] cpu_startup_entry+0xec/0x1e0 [ 134.259256] [<ffffffff817520b7>] rest_init+0x77/0x80 [ 134.259256] [<ffffffff81d230cf>] start_kernel+0x430/0x43d [ 134.259256] [<ffffffff81d22a8d>] ? set_init_arg+0x55/0x55 [ 134.259256] [<ffffffff81d22120>] ? early_idt_handler_array+0x120/0x120 [ 134.259256] [<ffffffff81d225d6>] x86_64_start_reservations+0x2a/0x2c [ 134.259256] [<ffffffff81d22715>] x86_64_start_kernel+0x13d/0x14c [ 134.259256] Code: 00 00 00 00 66 66 66 66 90 48 63 ff 55 48 c7 c0 3a a1 00 00 48 8b 14 fd c0 13 a7 81 48 89 e5 53 31 db 0f b7 0c 10 b8 05 00 00 00 <0f> 01 c1 5b 5d c3 0f 1f 00 66 66 66 66 90 55 65 48 8b 04 25 80 [ 134.259256] RIP [<ffffffff8104e8f7>] kvm_kick_cpu+0x27/0x30 [ 134.259256] RSP <ffff88107fc03df0> [ 134.259256] CR2: ffffffff8104e8f7 [ 0.000000] do_IRQ: 0.97 No irq handler for vector ÿdoing setup waiting for dump device vda5 dumping to vda5 Copying data : [100.0 %] / eta: 0s dump complete rebooting [ 18.867620] reboot: Restarting system Version-Release number of selected component (if applicable): Openshift 4.4.3 CNV 2.3 How reproducible: Sometime the guest OS runs fine but crash at other time. Steps to Reproduce: 1. Deploy the custom guest OS on CNV 2.3 2. Run the guest OS for a couple of days Actual results: Sometime cause a kernel panic Expected results: Guest OS runs fine
This is about supporting very old or custom guest OS kernels that are experiencing kernel panics due to an enabled pvspinlock. I've posted a PR to allow users to disable this option for their guests. https://github.com/kubevirt/kubevirt/pull/4972
verify with build virt-launcher-container-v2.6.2-4 step: 1. create fedora 12 vm 2. add below part in vm's yaml file to disable pvspinlock .... features: pvspinlock: enabled: false .... 3. start fedora12 vm and check libvirt xml ... <features> <acpi/> <pvspinlock state='off'/> </features> ... move to verified.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (OpenShift Virtualization 2.6.2 Images), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2021:1502
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 500 days