Same problem here. It freezes every month or under heavy load. KVM
running an an software RAID. Every month because there is an montly
cronjob to check the software Raid: /usr/share/mdadm/checkarray.
Testing the Partition, on which the kvm images are one it, triggers the
bug. The kvm images takes 100% cpu, only reboot can stop it:
syslog:
May 24 06:02:47 localhost kernel: [1682547.453843] INFO: task kdmflush:465
blocked for more than 120 seconds.
May 24 06:02:47 localhost kernel: [1682547.453845] echo 0
/proc/sys/kernel/hung_task_timeout_secs disables this message.
May 24 06:02:47 localhost kernel: [1682547.453848] kdmflush D
0 465 2 0x
May 24 06:02:47 localhost kernel: [1682547.453852] 88032e0b99d0
0046 00015dc0 00015dc0
May 24 06:02:47 localhost kernel: [1682547.453856] 88032ec8dfd0
88032e0b9fd8 00015dc0 88032ec8dc00
May 24 06:02:47 localhost kernel: [1682547.453860] 00015dc0
88032e0b9fd8 00015dc0 88032ec8dfd0
May 24 06:02:47 localhost kernel: [1682547.453864] Call Trace:
May 24 06:02:47 localhost kernel: [1682547.453879] [a0074685]
wait_barrier+0xf5/0x140 [raid1]
May 24 06:02:47 localhost kernel: [1682547.453885] [8105ded0] ?
default_wake_function+0x0/0x20
May 24 06:02:47 localhost kernel: [1682547.453890] [a0077651]
make_request+0x51/0x750 [raid1]
May 24 06:02:47 localhost kernel: [1682547.453894] [81064304] ?
check_preempt_wakeup+0x1c4/0x3c0
May 24 06:02:47 localhost kernel: [1682547.453897] [8105f10b] ?
enqueue_task_fair+0x9b/0xa0
May 24 06:02:47 localhost kernel: [1682547.453902] [8142b6b0]
md_make_request+0xc0/0x130
May 24 06:02:47 localhost kernel: [1682547.453907] [812a1d01]
generic_make_request+0x1b1/0x4f0
May 24 06:02:47 localhost kernel: [1682547.453911] [810f8475] ?
mempool_alloc_slab+0x15/0x20
May 24 06:02:47 localhost kernel: [1682547.453915] [810f860d] ?
mempool_alloc+0x5d/0x130
May 24 06:02:47 localhost kernel: [1682547.453919] [814382ad]
__map_bio+0xad/0x130
May 24 06:02:47 localhost kernel: [1682547.453922] [814387dd]
__clone_and_map+0x4ad/0x4c0
May 24 06:02:47 localhost kernel: [1682547.453925] [810f860d] ?
mempool_alloc+0x5d/0x130
May 24 06:02:47 localhost kernel: [1682547.453929] [814398b8]
__split_and_process_bio+0x108/0x190
May 24 06:02:47 localhost kernel: [1682547.453932] [81439996]
dm_flush+0x56/0x70
May 24 06:02:47 localhost kernel: [1682547.453935] [814399fc]
dm_wq_work+0x4c/0x1c0
May 24 06:02:47 localhost kernel: [1682547.453938] [814399b0] ?
dm_wq_work+0x0/0x1c0
May 24 06:02:47 localhost kernel: [1682547.453942] [81081457]
run_workqueue+0xc7/0x1a0
May 24 06:02:47 localhost kernel: [1682547.453946] [810815d3]
worker_thread+0xa3/0x110
May 24 06:02:47 localhost kernel: [1682547.453950] [81085ff0] ?
autoremove_wake_function+0x0/0x40
May 24 06:02:47 localhost kernel: [1682547.453954] [81081530] ?
worker_thread+0x0/0x110
May 24 06:02:47 localhost kernel: [1682547.453957] [81085c76]
kthread+0x96/0xa0
May 24 06:02:47 localhost kernel: [1682547.453961] [810141ea]
child_rip+0xa/0x20
May 24 06:02:47 localhost kernel: [1682547.453964] [81085be0] ?
kthread+0x0/0xa0
May 24 06:02:47 localhost kernel: [1682547.453967] [810141e0] ?
child_rip+0x0/0x20
May 24 06:02:47 localhost kernel: [1682547.453971] INFO: task jbd2/dm-0-8:610
blocked for more than 120 seconds.
May 24 06:02:47 localhost kernel: [1682547.453973] echo 0
/proc/sys/kernel/hung_task_timeout_secs disables this message.
May 24 06:02:47 localhost kernel: [1682547.453975] jbd2/dm-0-8 D
0 610 2 0x
May 24 06:02:47 localhost kernel: [1682547.453979] 880325db1d20
0046 00015dc0 00015dc0
May 24 06:02:47 localhost kernel: [1682547.453983] 8803265703d0
880325db1fd8 00015dc0 88032657
May 24 06:02:47 localhost kernel: [1682547.453986] 00015dc0
880325db1fd8 00015dc0 8803265703d0
May 24 06:02:47 localhost kernel: [1682547.453990] Call Trace:
May 24 06:02:47 localhost kernel: [1682547.453995] [8121e741]
jbd2_journal_commit_transaction+0x1c1/0x1280
May 24 06:02:47 localhost kernel: [1682547.453999] [81077bbc] ?
lock_timer_base+0x3c/0x70
May 24 06:02:47 localhost kernel: [1682547.454002] [81085ff0] ?
autoremove_wake_function+0x0/0x40
May 24 06:02:47 localhost kernel: [1682547.454006] [81225d7d]
kjournald2+0xbd/0x220
May 24 06:02:47 localhost kernel: [1682547.454010] [81085ff0] ?
autoremove_wake_function+0x0/0x40
May 24 06:02:47 localhost kernel: [1682547.454013] [81225cc0] ?
kjournald2+0x0/0x220
May 24 06:02:47 localhost kernel: [1682547.454016] [81085c76]
kthread+0x96/0xa0
May 24 06:02:47