View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000487 | AlmaLinux-8 | kernel | public | 2024-11-18 22:54 | 2024-11-18 22:54 |
| Reporter | Administratorte | Assigned To | |||
| Priority | high | Severity | crash | Reproducibility | always |
| Status | new | Resolution | open | ||
| Summary | 0000487: .22 and .27 kernels crash on two disperate PowerEdge Machines | ||||
| Description | Two machines, one old, one new: PowerEdge R520, Intel(R) Xeon(R) CPU E5-2430 v2 @ 2.50GHz PowerEdge R7525, AMD EPYC 7763 64-Core Processor Both run the .16 kernel fine. When the .22 kernel came out, I ran yum -y update on both machines and rebooted, but both machines crashed on reboot. The .22 and .27 kernels are fine on all other machines, including old HP desktops as well as new PowerEdge machines. One curious fact that may not be related is that on all other machines under /boot, there is always a corresponding initramfs*kdump.img to accompany every kernel, but on the failing machines these do not exist for the .22 and .27 kernels. They do exist for the .16 kernel and previous kernels. I tried the .27 kernel on the old machine with no change. I've yet to try the .27 kernel on the new machine as I suspect the behavior will not change. Capturing the crash turns out to be non-trivial, as it scrolls too fast and my attempts with the serial port have not worked. So far what I've been able to capture with a video camera: RIP: 0010(?)xhci_irq+0x140/0x3e0 (?) Code 00 0F 1F 40 00 48 03 e4 20 09 d0 5b 5d 41 5c 41 5d 41 5e 41 5b 01 00 00 00 0f 1f 44 00 00 0f ae e8 ? 4d 0c <89> c8 ... Call Trace <NMI> ? watchdog_overflow_callback.cold.7+0x1e/0x70 ? __perf_event_overflow+0x52/0x100 ? handle_pmi_common+0x200/0x2d0 ? __get_pto_vaddr+0x32/0x58 ?__native_set_fixmap+0x24/0x40 ? ghea(?)_copy_tofrom_phys+0xf9/0x250? ? intel_pmu_handle_irq+0x119/0x450 ? perf_event _nmi_handler+0x24/0x50 ? nmi_handle+0x63/0x110 ? default_do_nmi+0x19c/0x210 ? do_nmi+0x19c/0x69 ? end_repeat_nmi+0x16/0x69 ? xhci_irq+0x140/0x3e8 ? xhci_irq+0x140/0x3e8 </NMI> Later there is a kernel panic but the screen is blurry. I'll try to get better logs. | ||||
| Steps To Reproduce | yum -y update and reboot. .22 and .27 kernels crash, .16 is OK. | ||||
| Additional Information | no corresponding initramfs*kdump.img under /boot on the two machines that fail | ||||
| Tags | alma8, almalinux8, boot, kernel | ||||
| abrt_hash | |||||
| URL | |||||
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2024-11-18 22:54 | Administratorte | New Issue | |
| 2024-11-18 22:54 | Administratorte | Tag Attached: alma8 | |
| 2024-11-18 22:54 | Administratorte | Tag Attached: almalinux8 | |
| 2024-11-18 22:54 | Administratorte | Tag Attached: boot | |
| 2024-11-18 22:54 | Administratorte | Tag Attached: kernel |
