Comment 16 for bug 71567

Having the same issue..

Two separate hosts, runing Dapper, with LVM over MD, running Zimbra on a separate /opt LVM partition.
Crashes are sometimes hard panics, sometimes soft (dmesg shows a panic, some processes keep running, some get defunct,, ...)
Crash happens at the lvremove state, in a script simular to the ones posted above.

It got better (happens less) when syncing disks just before the lvremove, but it didn't stop happening. Sometimes the crash happens after the sync command and before the lvremove command.

One relmevant dmesg output after a soft panic:

[43045674.210000] Unable to handle kernel paging request at virtual address f8ab3004
[43045674.250000] printing eip:
[43045674.260000] f8aa84e2
[43045674.260000] *pde = 00000000
[43045674.280000] Oops: 0000 [#1]
[43045674.280000] SMP
[43045674.280000] Modules linked in: dm_snapshot usb_storage af_packet ppdev ipv6 dm_mod lp ide_floppy i2c_piix4 serio_raw parport_pc psmous
e parport i2c_core pcspkr floppy e100 sworks_agp mii agpgart sg evdev ext3 jbd raid1 md_mod ide_generic ehci_hcd ohci_hcd uhci_hcd usbcore 3
w_xxxx sd_mod aic7xxx scsi_transport_spi2 scsi_mod ide_cd cdrom serverworks generic thermal processor fan capability commoncap vga16fb vgast
ate fbcon tileblit font bitblit softcursor
[43045674.280000] CPU: 0
[43045674.280000] EIP: 0060:[<f8aa84e2>] Not tainted VLI
[43045674.280000] EFLAGS: 00010202 (2.6.15-29-server)
[43045674.280000] EIP is at persistent_commit+0x102/0x130 [dm_snapshot]
[43045674.280000] eax: 00000001 ebx: d20bf3e0 ecx: 00000000 edx: f8ab3000
[43045674.280000] esi: 00000001 edi: 00000000 ebp: f8aa6e00 esp: d11c7ed4
[43045674.280000] ds: 007b es: 007b ss: 0068
[43045674.280000] Process kcopyd (pid: 6553, threadinfo=d11c6000 task=ce7fba90)
[43045674.280000] Stack: d20bf3e0 00000001 00000001 0000120e 00000000 00000374 00000000 d205077c
[43045674.280000] 00000000 d357feb8 f8aa6e33 d2b990fc d357feb8 f8aa6df0 d357feb8 f8a7f793
[43045674.280000] 00000000 00000000 d357feb8 00000000 00000246 d205077c f8a86d60 00000000
[43045674.280000] Call Trace:
[43045674.280000] [<f8aa6e33>] copy_callback+0x33/0x50 [dm_snapshot]
[43045674.280000] [<f8aa6df0>] commit_callback+0x0/0x10 [dm_snapshot]
[43045674.280000] [<f8a7f793>] run_complete_job+0x63/0x80 [dm_mod]
[43045674.280000] [<f8a7f9b7>] process_jobs+0x17/0xe0 [dm_mod]
[43045674.280000] [<f8a7fa98>] do_work+0x18/0x50 [dm_mod]
[43045674.280000] [<f8a7f730>] run_complete_job+0x0/0x80 [dm_mod]
[43045674.280000] [<c0136fa3>] worker_thread+0x1b3/0x270
[43045674.280000] [<f8a7fa80>] do_work+0x0/0x50 [dm_mod]
[43045674.280000] [<c011f850>] default_wake_function+0x0/0x20
[43045674.280000] [<c0136df0>] worker_thread+0x0/0x270
[43045674.280000] [<c013bc18>] kthread+0xc8/0xd0
[43045674.280000] [<c013bb50>] kthread+0x0/0xd0
[43045674.280000] [<c0101505>] kernel_thread_helper+0x5/0x10
[43045674.280000] Code: c4 1c 5b 5e 5f c3 8d 76 00 8b 53 28 c7 43 08 00 00 00 00 85 d2 74 b7 31 f6 8b 43 2c 8d 14 f0 31 c0 85 ff 0f 94 c0 46
 89 44 24 04 <8b> 42 04 89 04 24 ff 12 39 73 28 77 e1 8b 43 10 39 43 20 c7 43
[43045674.280000]