Step to reproduce:
1. Create a VG with 2 disks.
2. Use lvcreate -m1 --mirrorlog mirrored -L 512M -n test vgTest
3. Use mkfs on the volume
4. Write some file and unplug one disk.
5. Unplug the other disk.
6. umount device, force removal using dmsetup remove -f on stuck devices.
7. dmsetup table shows nothing.
8. Plug back one disk.
9. Use vgchange -ay
10. Any LVM command will end in deadlock even after removing the disk.
Stack from dmesg:
[116167.589417] INFO: task vgchange:13916 blocked for more than 120 seconds.
[116167.589422] Tainted: G OE 3.16.0-38-generic #52~14.04.1-Ubuntu
[116167.589423] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[116167.589424] vgchange D ffff88032ee930c0 0 13916 11903 0x00000004
[116167.589428] ffff8800370539a0 0000000000000086 ffff88017fa1b2f0 ffff880037053fd8
[116167.589430] 00000000000130c0 00000000000130c0 ffff88012ae07010 ffff88032ee939c0
[116167.589432] ffff88017fa1b2f0 7fffffffffffffff ffff88017fa1b2f0 0000000000000000
[116167.589434] Call Trace:
[116167.589441] [<ffffffff817699f9>] schedule+0x29/0x70
[116167.589443] [<ffffffff81768dd9>] schedule_timeout+0x229/0x2a0
[116167.589448] [<ffffffff81355338>] ? submit_bio+0x78/0x160
[116167.589453] [<ffffffff8101bdb9>] ? read_tsc+0x9/0x20
[116167.589456] [<ffffffff810d9438>] ? ktime_get_ts+0x48/0xe0
[116167.589458] [<ffffffff8176a0cb>] io_schedule_timeout+0x9b/0xf0
[116167.589462] [<ffffffff815edd40>] ? bio_next_page+0x20/0x20
[116167.589465] [<ffffffff8176afe6>] wait_for_completion_io+0xa6/0x160
[116167.589468] [<ffffffff810a1550>] ? wake_up_state+0x20/0x20
[116167.589471] [<ffffffff815ee3d5>] sync_io+0xb5/0x110
[116167.589473] [<ffffffff81769dc9>] ? _cond_resched+0x29/0x40
[116167.589475] [<ffffffff8176a478>] ? wait_for_completion+0x28/0x160
[116167.589479] [<ffffffff8110149f>] ? __stop_cpus+0x4f/0x70
[116167.589481] [<ffffffff815ee60d>] dm_io+0x1dd/0x210
[116167.589483] [<ffffffff815edee0>] ? dm_io_client_destroy+0x30/0x30
[116167.589485] [<ffffffff815edd40>] ? bio_next_page+0x20/0x20
[116167.589489] [<ffffffffc0d6a92a>] disk_resume+0x15a/0x2e0 [dm_log]
[116167.589491] [<ffffffff810d448e>] ? synchronize_rcu_expedited+0xe/0x10
[116167.589493] [<ffffffffc0d80412>] mirror_resume+0x32/0x60 [dm_mirror]
[116167.589496] [<ffffffff815e9e0e>] dm_table_resume_targets+0x8e/0xe0
[116167.589499] [<ffffffff815e77cc>] dm_resume+0x4c/0xd0
[116167.589501] [<ffffffff815ec97b>] dev_suspend+0x12b/0x220
[116167.589503] [<ffffffff815ec850>] ? table_load+0x350/0x350
[116167.589505] [<ffffffff815ed215>] ctl_ioctl+0x255/0x500
[116167.589508] [<ffffffff812d7acf>] ? SYSC_semtimedop+0x23f/0xcf0
[116167.589511] [<ffffffff815ed4d3>] dm_ctl_ioctl+0x13/0x20
[116167.589515] [<ffffffff811e7360>] do_vfs_ioctl+0x2e0/0x4c0
[116167.589519] [<ffffffff812ec4e6>] ? security_sem_associate+0x16/0x20
[116167.589520] [<ffffffff812d6a89>] ? sem_security+0x9/0x10
[116167.589522] [<ffffffff812d4ee9>] ? ipcget+0x149/0x1c0
[116167.589524] [<ffffffff811e75c1>] SyS_ioctl+0x81/0xa0
[116167.589528] [<ffffffff8176da4d>] system_call_fastpath+0x1a/0x1f
[116167.589531] INFO: task lvs:14000 blocked for more than 120 seconds.
ProblemType: Bug
DistroRelease: Ubuntu 14.04
Package: linux-image-3.16.0-38-generic 3.16.0-38.52~14.04.1
ProcVersionSignature: Ubuntu 3.16.0-38.52~14.04.1-generic 3.16.7-ckt10
Uname: Linux 3.16.0-38-generic x86_64
ApportVersion: 2.14.1-0ubuntu3.11
Architecture: amd64
Date: Tue Jun 9 18:57:10 2015
ProcEnviron:
LANGUAGE=en_CA
TERM=xterm
PATH=(custom, no user)
LANG=en_CA.UTF-8
SHELL=/bin/bash
SourcePackage: linux-lts-utopic
UpgradeStatus: Upgraded to trusty on 2014-03-25 (441 days ago)
Jérôme Poulin, thank you for reporting this and helping make Ubuntu better. Please execute the following command only once, as it will automatically gather debugging information, in a terminal:
apport-collect 1463612