Bug#657690: General protection fault in xfs_vm_releasepage
Package: linux-image-2.6.32-5-amd64
Version: 2.6.32-39squeeze1
My backup server has begun crashing in the XFS code within two minutes
of the start of about one of every six AMANDA runs. The first crash,
twelve days ago, appears to be XFS bug 15516 and will be reported
separately (as I don't see an existing bug report). This morning's crash
follows:
Jan 27 00:11:22 chickenadventurous kernel: [811024.687095] general protection fault: 0000 [#1] SMP
Jan 27 00:11:22 chickenadventurous kernel: [811024.687142] last sysfs file: /sys/devices/pci0000:00/0000:00:1a.0/usb1/1-1/1-1.2/1-1.2:1.0/input/input7/event1/uevent
Jan 27 00:11:22 chickenadventurous kernel: [811024.687198] CPU 0
Jan 27 00:11:22 chickenadventurous kernel: [811024.687220] Modules linked in: rpcsec_gss_krb5 nfsd nfs lockd fscache nfs_acl auth_rpcgss sunrpc 8021q garp stp ext3 jbd mbcache loop snd_hda_codec_intelhdmi snd_hda_codec_realtek snd_hda_intel snd_hda_codec i915 snd_hwdep snd_pcm snd_timer snd drm_kms_helper drm psmouse i2c_algo_bit i2c_i801 pcspkr evdev soundcore video serio_raw snd_page_alloc i2c_core output button processor xfs exportfs dm_mod raid1 md_mod sg sr_mod cdrom osst sd_mod crc_t10dif st usbhid usb_storage hid ata_piix ata_generic sym53c8xx scsi_transport_spi e1000e libata ehci_hcd scsi_mod thermal usbcore nls_base thermal_sys [last unloaded: scsi_wait_scan]
Jan 27 00:11:22 chickenadventurous kernel: [811024.687541] Pid: 47, comm: kswapd0 Not tainted 2.6.32-5-amd64 #1
Jan 27 00:11:22 chickenadventurous kernel: [811024.687580] RIP: 0010:[<ffffffffa0206451>] [<ffffffffa0206451>] xfs_count_page_state+0x25/0x66 [xfs]
Jan 27 00:11:22 chickenadventurous kernel: [811024.687657] RSP: 0018:ffff880124867b78 EFLAGS: 00010203
Jan 27 00:11:22 chickenadventurous kernel: [811024.687692] RAX: ffff8801000000b0 RBX: ffffea0001728008 RCX: ffff880124867bd4
Jan 27 00:11:22 chickenadventurous kernel: [811024.687750] RDX: ffff880124867bd8 RSI: ffff880124867bdc RDI: ffffea0001728008
Jan 27 00:11:22 chickenadventurous kernel: [811024.687809] RBP: 0000000000000000 R08: 7a5da831f7287f78 R09: 6bc7c2d2d83dc935
Jan 27 00:11:22 chickenadventurous kernel: [811024.687869] R10: 0000000000000000 R11: ffffffffa0207a03 R12: ffff8800257a0d90
Jan 27 00:11:22 chickenadventurous kernel: [811024.687928] R13: 0000000000019e57 R14: 0000000000012805 R15: ffff880124867c40
Jan 27 00:11:22 chickenadventurous kernel: [811024.687988] FS: 0000000000000000(0000) GS:ffff880005200000(0000) knlGS:0000000000000000
Jan 27 00:11:22 chickenadventurous kernel: [811024.688048] CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
Jan 27 00:11:22 chickenadventurous kernel: [811024.688084] CR2: 00007fff89ae8a60 CR3: 0000000001001000 CR4: 00000000000006f0
Jan 27 00:11:22 chickenadventurous kernel: [811024.688142] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Jan 27 00:11:22 chickenadventurous kernel: [811024.688202] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Jan 27 00:11:22 chickenadventurous kernel: [811024.688258] Process kswapd0 (pid: 47, threadinfo ffff880124866000, task ffff880127b9dbd0)
Jan 27 00:11:22 chickenadventurous kernel: [811024.688320] Stack:
Jan 27 00:11:22 chickenadventurous kernel: [811024.688346] ffffffffa0207a53 0000000000000000 0000000000000000 0000000000000001
Jan 27 00:11:22 chickenadventurous kernel: [811024.688384] <0> 0000000000000000 0000000000000001 0000000000000000 0000000000000000
Jan 27 00:11:22 chickenadventurous kernel: [811024.688448] <0> 0000000000000000 0000000000000000 ffffea0001728040 0000000101728040
Jan 27 00:11:22 chickenadventurous kernel: [811024.690605] Call Trace:
Jan 27 00:11:22 chickenadventurous kernel: [811024.690639] [<ffffffffa0207a53>] ? xfs_vm_releasepage+0x50/0xa5 [xfs]
Jan 27 00:11:22 chickenadventurous kernel: [811024.690683] [<ffffffff810bcf51>] ? invalidate_inode_page+0x5c/0x87
Jan 27 00:11:22 chickenadventurous kernel: [811024.690721] [<ffffffff810bcfe5>] ? invalidate_mapping_pages+0x69/0xdb
Jan 27 00:11:22 chickenadventurous kernel: [811024.690759] [<ffffffff81100817>] ? shrink_icache_memory+0xfc/0x228
Jan 27 00:11:22 chickenadventurous kernel: [811024.690797] [<ffffffff810bf639>] ? shrink_slab+0xe0/0x153
Jan 27 00:11:22 chickenadventurous kernel: [811024.690833] [<ffffffff810bfedc>] ? kswapd+0x4d9/0x686
Jan 27 00:11:22 chickenadventurous kernel: [811024.690867] [<ffffffff810bd553>] ? isolate_pages_global+0x0/0x20f
Jan 27 00:11:22 chickenadventurous kernel: [811024.690907] [<ffffffff81064f92>] ? autoremove_wake_function+0x0/0x2e
Jan 27 00:11:22 chickenadventurous kernel: [811024.690947] [<ffffffff8103aa66>] ? __wake_up_common+0x44/0x72
Jan 27 00:11:22 chickenadventurous kernel: [811024.690985] [<ffffffff810bfa03>] ? kswapd+0x0/0x686
Jan 27 00:11:22 chickenadventurous kernel: [811024.691017] [<ffffffff81064cc5>] ? kthread+0x79/0x81
Jan 27 00:11:22 chickenadventurous kernel: [811024.691052] [<ffffffff81011baa>] ? child_rip+0xa/0x20
Jan 27 00:11:22 chickenadventurous kernel: [811024.691087] [<ffffffff81064c4c>] ? kthread+0x0/0x81
Jan 27 00:11:22 chickenadventurous kernel: [811024.691119] [<ffffffff81011ba0>] ? child_rip+0x0/0x20
Jan 27 00:11:22 chickenadventurous kernel: [811024.691153] Code: 49 89 1f eb ef 90 c7 01 00 00 00 00 c7 02 00 00 00 00 c7 06 00 00 00 00 48 8b 07 f6 c4 08 75 04 0f 0b eb fe 48 8b 47 10 49 89 c1 <4d> 8b 01 41 f6 c0 01 74 0e 41 f6 c0 20 75 08 c7 02 01 00 00 00
Jan 27 00:11:22 chickenadventurous kernel: [811024.691301] RIP [<ffffffffa0206451>] xfs_count_page_state+0x25/0x66 [xfs]
Jan 27 00:11:22 chickenadventurous kernel: [811024.691347] RSP <ffff880124867b78>
Jan 27 00:11:22 chickenadventurous kernel: [811024.691556] ---[ end trace 6ac7a1d4e5f740a4 ]---
The system was responsive enough to run 'dmesg', 'dpkg -l', 'grep' and
'less' on an existing SSH connection, but the session became unusable
after I tried to use 'sudo'. My NRPE check plugins that use Perl aren't
responding, but other checks report 649 processes (twelve zombies) and a
load average of 389.75. Remote syslog is still working and apparently
cron jobs are firing off.
If memory serves (the system is still down; I need to fix an unrelated
problem first), all filesystems are XFS on LVM on Linux software RAID,
except one which is either ext3 or xfs on a USB flash drive.
--
William Aoki KD7YAF waoki@umnh.utah.edu 5-1924
Reply to: