Утро добрым не бывает. Снова лёг сервер с Oracle Linux на борту, уже другой. Ошибка:
general protection fault: 0000 [#1] SMP
Сервер ушёл в перезагрузку.
Окружение
HP ProLiant DL580 Gen9 Oracle Linux Server release 7.6 4.1.12-124.22.4.el7uek.x86_64
-
Лог
[487447.471224] general protection fault: 0000 [#1] SMP [487447.471316] Modules linked in: tcp_diag udp_diag inet_diag unix_diag af_packet_diag netlink_diag arc4 ecb md4 nls_utf8 cifs dns_resolver bonding sunrpc xfs vfat fat iTCO_wdt iTCO_vendor_support intel_powerclamp coretemp raid1 kvm_intel kvm raid10 crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd pcspkr dm_multipath sb_edac ipmi_ssif sg edac_core dm_mod hpilo hpwdt lpc_ich mfd_core shpchp wmi ipmi_si ipmi_msghandler acpi_cpufreq binfmt_misc ip_tables ext4 mbcache2 jbd2 sd_mod mgag200 syscopyarea sysfillrect sysimgblt i2c_algo_bit drm_kms_helper ttm drm i2c_core bnx2x mdio libcrc32c nvme crc32c_intel hpsa ptp nvme_core pps_core scsi_transport_sas [487447.471995] CPU: 88 PID: 0 Comm: swapper/88 Not tainted 4.1.12-124.22.4.el7uek.x86_64 #2 [487447.472059] Hardware name: HP ProLiant DL580 Gen9/ProLiant DL580 Gen9, BIOS U17 10/11/2018 [487447.472124] task: ffff884f0b0d8000 ti: ffff884f0b0e0000 task.ti: ffff884f0b0e0000 [487447.472182] RIP: 0010:[<ffffffff81372d60>] [<ffffffff81372d60>] swiotlb_unmap_sg_attrs+0x30/0x70 [487447.472263] RSP: 0018:ffff884f3ec03d98 EFLAGS: 00010087 [487447.472307] RAX: 0005516700b0ebc8 RBX: 0005516700b0ebc8 RCX: 0000000000000001 [487447.472362] RDX: 0005516700b0ebc9 RSI: 00000002c0fe4000 RDI: ffff88163cc3e9a0 [487447.472417] RBP: ffff884f3ec03dc8 R08: 0000000000000000 R09: ffffffff81372d30 [487447.472473] R10: 0000000000000000 R11: 0000000000000000 R12: 000000000000004e [487447.472528] R13: 00000000000000c0 R14: 0000000000000001 R15: ffff884f07953098 [487447.472585] FS: 0000000000000000(0000) GS:ffff884f3ec00000(0000) knlGS:0000000000000000 [487447.472647] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [487447.472693] CR2: 00007f7ababf9fb0 CR3: 0000000001b46000 CR4: 0000000000360670 [487447.472749] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [487447.472804] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [487447.472858] Stack: [487447.472879] ffff884f3ec03db8 ffff880061d09000 ffff8801006f0c40 ffff884040458000 [487447.472946] ffff880061ce0900 ffff883fa8b88000 ffff884f3ec03dd8 ffffffff814e4108 [487447.473012] ffff884f3ec03e48 ffffffffa00492d6 ffff884f3ec03e18 ffffffff810fa997 [487447.473078] Call Trace: [487447.473102] <IRQ> [487447.473130] [<ffffffff814e4108>] scsi_dma_unmap+0x58/0x70 [487447.473188] [<ffffffffa00492d6>] finish_cmd+0x96/0x940 [hpsa] [487447.473241] [<ffffffff810fa997>] ? hrtimer_get_next_event+0x47/0xa0 [487447.473299] [<ffffffffa004c229>] do_hpsa_intr_msi+0xa9/0x160 [hpsa] [487447.475282] [<ffffffff810e7db0>] handle_irq_event_percpu+0x60/0x210 [487447.477201] [<ffffffff810e7fa1>] handle_irq_event+0x41/0x70 [487447.479097] [<ffffffff810eb46f>] handle_edge_irq+0x7f/0x140 [487447.481044] [<ffffffff8101a634>] handle_irq+0xb4/0x140 [487447.482872] [<ffffffff810ad73a>] ? atomic_notifier_call_chain+0x1a/0x20 [487447.484668] [<ffffffff8175b211>] do_IRQ+0x51/0x100 [487447.486454] [<ffffffff81754e56>] common_interrupt+0x196/0x196 [487447.488236] <EOI> [487447.488260] [<ffffffff815d2f21>] ? cpuidle_enter_state+0xd1/0x240 [487447.491835] [<ffffffff815d2ef1>] ? cpuidle_enter_state+0xa1/0x240 [487447.493612] [<ffffffff815d30c7>] cpuidle_enter+0x17/0x20 [487447.495380] [<ffffffff810d29a6>] cpu_startup_entry+0x236/0x320 [487447.497149] [<ffffffff810563a5>] start_secondary+0x1a5/0x210 [487447.498884] Code: 57 41 56 41 89 ce 41 55 41 54 53 48 83 ec 08 83 f9 03 74 4c 45 31 e4 85 d2 49 89 ff 48 89 f3 41 89 d5 7e 2d 0f 1f 80 00 00 00 00 <8b> 53 18 48 8b 73 10 44 89 f1 4c 89 ff 41 83 c4 01 e8 7a ff ff [487447.502664] RIP [<ffffffff81372d60>] swiotlb_unmap_sg_attrs+0x30/0x70 [487447.504461] RSP <ffff884f3ec03d98>
Особо ничего путного в Интернет не нашёл, ошибка без особых подробностей. Судя по словам "scsi_dma_unmap", "do_hpsa_intr_msi", "hpsa" проблема где-то в драйверах HPE, но это не точно.
Поищу новые драйвера. Если не поможет: обновлять ОС.
