Hi WD, Forum Fellas,
Some time ago, I wrote an article where I described how to fix the MBWE white lights periodic hang:
http://mybookworld.wikidot.com/forum/t-284602/mbwe-2t-does-not-mount-directories
That worked for 6+ months to me.
OTOH, a couple of days ago my MBWE was hung again. And I performed the steps described on the article. And at this time, it didn’t help. Now the drive keep hanging when under heavy load.
Meanwhile, I managed to hang the filesys while logged in the MBWE, and I was able to get a kernel thread dump when the filesystem withdrew:
XFS internal error XFS_WANT_CORRUPTED_GOTO at line 2058 of file fs/xfs/xfs_bmap_btree.c. Caller 0xc0110820
[<c0028e30>] (dump_stack+0x0/0x14) from [<c012f104>] (xfs_error_report+0x54/0x64)
[<c012f0b0>] (xfs_error_report+0x0/0x64) from [<c011e558>] (xfs_bmbt_insert+0x170/0x184)
r4:c7f41f48
[<c011e3e8>] (xfs_bmbt_insert+0x0/0x184) from [<c0110820>] (xfs_bmap_add_extent_unwritten_real+0xe0c/0xe6c)
[<c010fa14>] (xfs_bmap_add_extent_unwritten_real+0x0/0xe6c) from [<c0115780>] (xfs_bmap_add_extent+0x134/0x514)
[<c011564c>] (xfs_bmap_add_extent+0x0/0x514) from [<c01179a0>] (xfs_bmapi+0xeb0/0x139c)
[<c0116af0>] (xfs_bmapi+0x0/0x139c) from [<c013e2c0>] (xfs_iomap_write_unwritten+0x170/0x250)
[<c013e150>] (xfs_iomap_write_unwritten+0x0/0x250) from [<c013e740>] (xfs_iomap+0x2c8/0x4bc)
[<c013e478>] (xfs_iomap+0x0/0x4bc) from [<c0166360>] (xfs_bmap+0x30/0x38)
[<c0166330>] (xfs_bmap+0x0/0x38) from [<c015db00>] (xfs_end_bio_unwritten+0x48/0x60)
[<c015dab8>] (xfs_end_bio_unwritten+0x0/0x60) from [<c0046314>] (run_workqueue+0xec/0x1a0)
r4:c7c77b60
[<c0046228>] (run_workqueue+0x0/0x1a0) from [<c0046e10>] (worker_thread+0x78/0xd0)
r6:c7ca7fac r5:c7c77b68 r4:c7c77b60
[<c0046d98>] (worker_thread+0x0/0xd0) from [<c004a59c>] (kthread+0x50/0x7c)
r6:00000000 r5:c0046d98 r4:c7c77b60
[<c004a54c>] (kthread+0x0/0x7c) from [<c0038e40>] (do_exit+0x0/0x7ec)
r5:00000000 r4:00000000
Filesystem "md2": XFS internal error xfs_trans_cancel at line 1163 of file fs/xfs/xfs_trans.c. Caller 0xc013e378
[<c0028e30>] (dump_stack+0x0/0x14) from [<c012f104>] (xfs_error_report+0x54/0x64)
[<c012f0b0>] (xfs_error_report+0x0/0x64) from [<c0150f30>] (xfs_trans_cancel+0x108/0x130)
r4:0044c008
[<c0150e28>] (xfs_trans_cancel+0x0/0x130) from [<c013e378>] (xfs_iomap_write_unwritten+0x228/0x250)
r8:00010ece r7:00000000 r6:00000011 r5:00000004 r4:00000075
[<c013e150>] (xfs_iomap_write_unwritten+0x0/0x250) from [<c013e740>] (xfs_iomap+0x2c8/0x4bc)
[<c013e478>] (xfs_iomap+0x0/0x4bc) from [<c0166360>] (xfs_bmap+0x30/0x38)
[<c0166330>] (xfs_bmap+0x0/0x38) from [<c015db00>] (xfs_end_bio_unwritten+0x48/0x60)
[<c015dab8>] (xfs_end_bio_unwritten+0x0/0x60) from [<c0046314>] (run_workqueue+0xec/0x1a0)
r4:c7c77b60
[<c0046228>] (run_workqueue+0x0/0x1a0) from [<c0046e10>] (worker_thread+0x78/0xd0)
r6:c7ca7fac r5:c7c77b68 r4:c7c77b60
[<c0046d98>] (worker_thread+0x0/0xd0) from [<c004a59c>] (kthread+0x50/0x7c)
r6:00000000 r5:c0046d98 r4:c7c77b60
[<c004a54c>] (kthread+0x0/0x7c) from [<c0038e40>] (do_exit+0x0/0x7ec)
r5:00000000 r4:00000000
xfs_force_shutdown(md2,0x8) called from line 1164 of file fs/xfs/xfs_trans.c. Return address = 0xc0150f44
Filesystem "md2": Corruption of in-memory data detected. Shutting down filesystem: md2
Please umount the filesystem, and rectify the problem(s)
This issue was linked to kernel patch at this URL: https://patchwork.kernel.org/patch/3288/
Hey WD, any chances to brew us a new kernel containing this one-line fix?
Needless to say it will fix further occurences of XFS hang and save you extra/further headache.