Le Tue, 4 Apr 2017 15:02:02 +0200
Emmanuel Florac <eflorac@xxxxxxxxxxxxxx> écrivait:
> Hi,
> here is an interesting crash dump. I've never seen this function
> called before : "Fixing recursive fault but reboot is needed!"
>
> The system is frozen and must be hard rebooted. The filesystem is
> humongous: about 400 TB.
>
> Running is plain vanilla 3.18.44 (should upgrade...). I wonder if the
> bug is still present though?
>
> Context: this is a file server. There was a power failure yesterday,
> so there's probably some corruption hidden somewhere triggering the
> crash.
>
The machine goes on crashing on disk access... xfs_repair 4.9 is
running now, and after that we'll reboot with a current kernel
(4.4.59). Any advice?
The latest crash trace was still xfs related apparently:
avril 04 16:22:01 Colorstock-01 kernel: BUG: unable to handle kernel NULL pointer dereference at 0000000000000240
avril 04 16:22:01 Colorstock-01 kernel: IP: [<ffffffff811614d5>] iput+0x5/0x190
avril 04 16:22:01 Colorstock-01 kernel: PGD 0
avril 04 16:22:01 Colorstock-01 kernel: Oops: 0000 [#1] SMP
avril 04 16:22:01 Colorstock-01 kernel: Modules linked in: nfsv3 arc4 ecb md4 nfsv4 cifs fscache dm_mod cfg80211 rfkill nfsd auth_rpcgss oid_registry nfs_acl nfs lockd grace sunrpc af_packet bonding joydev evdev x86_pkg_temp_thermal coretemp kvm_intel kvm crct10dif_pclmul crc32_pclmul crc32c_intel aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd microcode pcspkr myri10ge ast ttm drm_kms_helper drm i2c_algo_bit ixgbe ptp pps_core mdio dca ses enclosure i2c_i801 sg i2c_core lpc_ich mfd_core ipmi_si 8250_fintek ipmi_msghandler rtc_cmos wmi processor thermal_sys acpi_power_meter button md_mod usbhid xhci_pci xhci_hcd ohci_pci ehci_pci ohci_hcd uhci_hcd ehci_hcd uas usb_storage usbcore usb_common fuse ipv6 autofs4 ext4 ext3 xfs reiserfs crc16 jbd2 jbd aacraid ahci libahci ata_generic libata
avril 04 16:22:01 Colorstock-01 kernel: CPU: 19 PID: 822 Comm: kworker/u64:11 Not tainted 3.18.44-storiq64-i7 #1
avril 04 16:22:01 Colorstock-01 kernel: Hardware name: Supermicro Super Server/X10DRD-iNT, BIOS 2.0 12/17/2015
avril 04 16:22:01 Colorstock-01 kernel: Workqueue: writeback bdi_writeback_workfn (flush-253:0)
avril 04 16:22:01 Colorstock-01 kernel: task: ffff88085c19a050 ti: ffff88085c1ac000 task.ti: ffff88085c1ac000
avril 04 16:22:01 Colorstock-01 kernel: RIP: 0010:[<ffffffff811614d5>] [<ffffffff811614d5>] iput+0x5/0x190
avril 04 16:22:01 Colorstock-01 kernel: RSP: 0018:ffff88085c1af590 EFLAGS: 00010202
avril 04 16:22:01 Colorstock-01 kernel: RAX: 0000000000000000 RBX: ffff880755ea4000 RCX: 000ffffffffe0000
avril 04 16:22:01 Colorstock-01 kernel: RDX: 0000000000000000 RSI: 0000000000000060 RDI: 00000000000001a8
avril 04 16:22:01 Colorstock-01 kernel: RBP: 0000000000000000 R08: ffff88085c1af684 R09: ffff88085c1af750
avril 04 16:22:01 Colorstock-01 kernel: R10: 0000000000000001 R11: 0000000000000000 R12: ffff880652e45c00
avril 04 16:22:01 Colorstock-01 kernel: R13: 0000000000000000 R14: ffff88085c1af801 R15: ffff880755ea4000
avril 04 16:22:01 Colorstock-01 kernel: FS: 0000000000000000(0000) GS:ffff88087f460000(0000) knlGS:0000000000000000
avril 04 16:22:01 Colorstock-01 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
avril 04 16:22:01 Colorstock-01 kernel: CR2: 0000000000000240 CR3: 00000000015e9000 CR4: 00000000003407e0
avril 04 16:22:01 Colorstock-01 kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
avril 04 16:22:01 Colorstock-01 kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
avril 04 16:22:01 Colorstock-01 kernel: Stack:
avril 04 16:22:01 Colorstock-01 kernel: ffffffffa2c357b6 0000000000001000 ffffffff52e45c60 ffff880652e45c40
avril 04 16:22:01 Colorstock-01 kernel: ffff88085c1af730 ffff88085c1af730 0000000000000010 0000000000000000
avril 04 16:22:01 Colorstock-01 kernel: ffffffffa2c0438d ffff88085c8c6c48 ffff88085c1af770 ffff88085c1af7b0
avril 04 16:22:01 Colorstock-01 kernel: Call Trace:
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c357b6>] ? xfs_filestream_lookup_ag+0x76/0x1b0 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c0438d>] ? xfs_bmap_btalloc+0x2dd/0x770 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2bfc192>] ? xfs_bmap_search_multi_extents+0xa2/0x120 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c03a06>] ? xfs_bmap_last_extent+0x56/0x80 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c03a56>] ? xfs_bmap_isaeof+0x26/0x90 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c05211>] ? xfs_bmapi_write+0x4b1/0xaa0 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c3ca61>] ? xfs_iomap_write_allocate+0x131/0x350 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c283d7>] ? xfs_map_blocks+0x1b7/0x240 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffffa2c291f6>] ? xfs_vm_writepage+0x186/0x5a0 [xfs]
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81106ead>] ? __writepage+0xd/0x40
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff811077af>] ? write_cache_pages+0x1bf/0x430
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81106ea0>] ? global_dirtyable_memory+0x50/0x50
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81107a58>] ? generic_writepages+0x38/0x50
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff8116db61>] ? __writeback_single_inode+0x41/0x2b0
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff811700b8>] ? writeback_sb_inodes+0x1b8/0x470
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff811703fe>] ? __writeback_inodes_wb+0x8e/0xc0
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81170683>] ? wb_writeback+0x253/0x350
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81170b8a>] ? bdi_writeback_workfn+0x2ca/0x480
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81071b95>] ? process_one_work+0x155/0x440
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81072543>] ? worker_thread+0x63/0x490
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff810724e0>] ? rescuer_thread+0x290/0x290
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff810769ee>] ? kthread+0xce/0xf0
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81076920>] ? kthread_create_on_node+0x180/0x180
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81450fd8>] ? ret_from_fork+0x58/0x90
avril 04 16:22:01 Colorstock-01 kernel: [<ffffffff81076920>] ? kthread_create_on_node+0x180/0x180
avril 04 16:22:01 Colorstock-01 kernel: Code: 08 48 8d b8 00 04 00 00 e8 e9 eb fb ff 84 c0 74 09 65 48 ff 04 25 18 f5 00 00 48 83 c4 08 c3 0f 1f 80 00 00 00 00 48 85 ff 74 3e <f6> 87 98 00 00 00 40 0f 85 fe 00 00 00 41 55 41 54 55 48 8d af
avril 04 16:22:01 Colorstock-01 kernel: RIP [<ffffffff811614d5>] iput+0x5/0x190
avril 04 16:22:01 Colorstock-01 kernel: RSP <ffff88085c1af590>
avril 04 16:22:01 Colorstock-01 kernel: CR2: 0000000000000240
avril 04 16:22:01 Colorstock-01 kernel: ---[ end trace a6342d7a0e4dea0f ]---
avril 04 16:22:01 Colorstock-01 kernel: ------------[ cut here ]------------
--
------------------------------------------------------------------------
Emmanuel Florac | Direction technique
| Intellique
| <eflorac@xxxxxxxxxxxxxx>
| +33 1 78 94 84 02
------------------------------------------------------------------------
Attachment:
pgpS4EDGljctt.pgp
Description: Signature digitale OpenPGP
