[Xen-devel] drdb fails under xen

[Luciano Miguel Ferreira Rocha <strange@xxxxxxxxxxxxx>]

Under xen, either stock xen-2.0.1+linux-2.6.9 or Fedora's latest
(2.6.10-1.1109_FC4), drdb (<http://www.drbd.org/>),as soon as I try to
write to one of its devices, causes "Badness" in the kernel and,
ultimately, a spontaneous reboot.

Under non-xen 2.6.10-1.1109_FC4 and 2.6.9-1.678_FC3, drbd works fine.

The error:
Jan 25 21:01:23 turing kernel: Badness in blk_plug_device at 
drivers/block/ll_rw_blk.c:1217
Jan 25 21:01:23 turing kernel:  [<c022942a>] blk_plug_device+0x2e/0x65
Jan 25 21:01:23 turing kernel:  [<c8985884>] 
drbd_make_request_common+0x4a6/0x8b9 [drbd]
Jan 25 21:01:23 turing kernel:  [<c016abea>] do_open+0xc5/0x2de
Jan 25 21:01:23 turing kernel:  [<c0146e40>] poison_obj+0x21/0x41
Jan 25 21:01:23 turing kernel:  [<c014872b>] 
cache_alloc_debugcheck_after+0xe8/0xf0
Jan 25 21:01:23 turing kernel:  [<c8985e5e>] drbd_make_request_26+0x1c7/0x1cf 
[drbd]
Jan 25 21:01:23 turing kernel:  [<c022b591>] generic_make_request+0x1a3/0x1b5
Jan 25 21:01:23 turing kernel:  [<c013476c>] autoremove_wake_function+0x0/0x3a
Jan 25 21:01:23 turing kernel:  [<c0142888>] mempool_alloc_slab+0xd/0x10
Jan 25 21:01:23 turing kernel:  [<c013476c>] autoremove_wake_function+0x0/0x3a
Jan 25 21:01:23 turing kernel:  [<c013476c>] autoremove_wake_function+0x0/0x3a
Jan 25 21:01:23 turing kernel:  [<c022b638>] submit_bio+0x95/0x9d
Jan 25 21:01:23 turing kernel:  [<c0166b05>] bio_alloc+0x105/0x170
Jan 25 21:01:23 turing kernel:  [<c0166481>] submit_bh+0x124/0x14a
Jan 25 21:01:23 turing kernel:  [<c01649e9>] __bread_slow+0x4a/0x86
Jan 25 21:01:23 turing kernel:  [<c0164c06>] __bread+0x26/0x2c
Jan 25 21:01:23 turing kernel:  [<c885d0ef>] ext3_fill_super+0xcd/0x9c1 [ext3]
Jan 25 21:01:23 turing kernel:  [<c01a0a89>] disk_name+0x25/0x66
Jan 25 21:01:23 turing kernel:  [<c0169633>] get_sb_bdev+0xdc/0x126
Jan 25 21:01:23 turing kernel:  [<c0140030>] do_generic_mapping_read+0x262/0x3d9
Jan 25 21:01:23 turing kernel:  [<c01802c3>] alloc_vfsmnt+0x7f/0xa8
Jan 25 21:01:23 turing kernel:  [<c885eb55>] ext3_get_sb+0x1a/0x1e [ext3]
Jan 25 21:01:23 turing kernel:  [<c885d022>] ext3_fill_super+0x0/0x9c1 [ext3]
Jan 25 21:01:23 turing kernel:  [<c0169828>] do_kern_mount+0x85/0x12c
Jan 25 21:01:24 turing kernel:  [<c0181baf>] do_new_mount+0x67/0xa4
Jan 25 21:01:24 turing kernel:  [<c0182514>] do_mount+0x15f/0x179
Jan 25 21:01:24 turing kernel:  [<c018236a>] copy_mount_options+0x49/0x94
Jan 25 21:01:24 turing kernel:  [<c0182936>] sys_mount+0x78/0xce
Jan 25 21:01:24 turing kernel:  [<c0109cac>] syscall_call+0x7/0xb
Jan 25 21:01:24 turing kernel: Badness in blk_remove_plug at 
drivers/block/ll_rw_blk.c:1238
Jan 25 21:01:24 turing kernel:  [<c022948f>] blk_remove_plug+0x2e/0x57
Jan 25 21:01:24 turing kernel:  [<c898cdfc>] drbd_unplug_fn+0x7c/0x353 [drbd]
Jan 25 21:01:24 turing kernel:  [<c0162e1e>] sync_buffer+0x2b/0x35
Jan 25 21:01:24 turing kernel:  [<c02ec013>] __wait_on_bit+0x2b/0x51
Jan 25 21:01:24 turing kernel:  [<c02ec0a7>] out_of_line_wait_on_bit+0x6e/0x76
Jan 25 21:01:24 turing kernel:  [<c0162df3>] sync_buffer+0x0/0x35
Jan 25 21:01:24 turing kernel:  [<c01347a6>] wake_bit_function+0x0/0x34
Jan 25 21:01:24 turing kernel:  [<c0166481>] submit_bh+0x124/0x14a
Jan 25 21:01:24 turing kernel:  [<c01347a6>] wake_bit_function+0x0/0x34
Jan 25 21:01:24 turing kernel:  [<c0164a0f>] __bread_slow+0x70/0x86
Jan 25 21:01:24 turing kernel:  [<c0164c06>] __bread+0x26/0x2c
Jan 25 21:01:24 turing kernel:  [<c885d0ef>] ext3_fill_super+0xcd/0x9c1 [ext3]
Jan 25 21:01:24 turing kernel:  [<c01a0a89>] disk_name+0x25/0x66
Jan 25 21:01:24 turing kernel:  [<c0169633>] get_sb_bdev+0xdc/0x126
Jan 25 21:01:24 turing kernel:  [<c0140030>] do_generic_mapping_read+0x262/0x3d9
Jan 25 21:01:24 turing kernel:  [<c01802c3>] alloc_vfsmnt+0x7f/0xa8
Jan 25 21:01:24 turing kernel:  [<c885eb55>] ext3_get_sb+0x1a/0x1e [ext3]
Jan 25 21:01:24 turing kernel:  [<c885d022>] ext3_fill_super+0x0/0x9c1 [ext3]
Jan 25 21:01:24 turing kernel:  [<c0169828>] do_kern_mount+0x85/0x12c
Jan 25 21:01:25 turing kernel:  [<c0181baf>] do_new_mount+0x67/0xa4
Jan 25 21:01:25 turing kernel:  [<c0182514>] do_mount+0x15f/0x179
Jan 25 21:01:25 turing kernel:  [<c018236a>] copy_mount_options+0x49/0x94
Jan 25 21:01:25 turing kernel:  [<c0182936>] sys_mount+0x78/0xce
Jan 25 21:01:25 turing kernel:  [<c0109cac>] syscall_call+0x7/0xb

and keeps repeating itself.

It works fine without xen, but fails under xen. Could xen be catching a
bug on drbd, or is it a bug caused by xen's paches?

Regards,
Luciano Rocha

-- 
2/16


-------------------------------------------------------
This SF.Net email is sponsored by: IntelliVIEW -- Interactive Reporting
Tool for open source databases. Create drag-&-drop reports. Save time
by over 75%! Publish reports on the web. Export to DOC, XLS, RTF, etc.
Download a FREE copy at http://www.intelliview.com/go/osdn_nl
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/xen-devel