mount_ubifs: failed to mount ubifs

mount_ubifs: failed to mount ubifs

[Bean Huo 霍斌斌 (beanhuo)]

Hello:
Currently ,I mainly focus on ubi powerloss solution for MLC nand ,now I 
Have a problem about master node, below is log.

When PEB 354:4096 read failed,and ubi will regard this block a corrupted block.
Ubi will not recover it, but reserve it.
Maybe this block contains master node, so will mount failure?
Why ubi not try to recover such block?



UBI: attaching mtd2 to ubi0 ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI error: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read 4096 bytes
CPU: 1 PID: 1303 Comm: ubiattach Not tainted 3.14.0-xilinx-00017-g71d29dc-dirty #65
[<c00145c0>] (unwind_backtrace) from [<c0010c50>] (show_stack+0x10/0x14)
[<c0010c50>] (show_stack) from [<c0478df4>] (dump_stack+0x84/0xd4)
[<c0478df4>] (dump_stack) from [<c02bacf0>] (ubi_io_read+0x1c4/0x298)
[<c02bacf0>] (ubi_io_read) from [<c02c1338>] (ubi_backup_volume_scan+0x208/0x308)
[<c02c1338>] (ubi_backup_volume_scan) from [<c02bf300>] (ubi_attach+0x29c/0x1200)
[<c02bf300>] (ubi_attach) from [<c02b60a0>] (ubi_attach_mtd_dev+0x59c/0xb60)
[<c02b60a0>] (ubi_attach_mtd_dev) from [<c02b68b4>] (ctrl_cdev_ioctl+0xd0/0x17c)
[<c02b68b4>] (ctrl_cdev_ioctl) from [<c00bfac4>] (vfs_ioctl+0x18/0x34)
[<c00bfac4>] (vfs_ioctl) from [<c00c059c>] (do_vfs_ioctl+0x530/0x578)
[<c00c059c>] (do_vfs_ioctl) from [<c00c061c>] (SyS_ioctl+0x38/0x54)
[<c00c061c>] (SyS_ioctl) from [<c000db80>] (ret_fast_syscall+0x0/0x30)
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI warning: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read only 4096 bytes, retry
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
ecc unrecoverable error
UBI error: ubi_io_read: error -74 (ECC error) while reading 4096 bytes from PEB 354:4096, read 4096 bytes
CPU: 1 PID: 1303 Comm: ubiattach Not tainted 3.14.0-xilinx-00017-g71d29dc-dirty #65
[<c00145c0>] (unwind_backtrace) from [<c0010c50>] (show_stack+0x10/0x14)
[<c0010c50>] (show_stack) from [<c0478df4>] (dump_stack+0x84/0xd4)
[<c0478df4>] (dump_stack) from [<c02bacf0>] (ubi_io_read+0x1c4/0x298)
[<c02bacf0>] (ubi_io_read) from [<c02bb1a0>] (ubi_io_read_vid_hdr+0x6c/0x1d4)
[<c02bb1a0>] (ubi_io_read_vid_hdr) from [<c02bf2ec>] (ubi_attach+0x288/0x1200)
[<c02bf2ec>] (ubi_attach) from [<c02b60a0>] (ubi_attach_mtd_dev+0x59c/0xb60)
[<c02b60a0>] (ubi_attach_mtd_dev) from [<c02b68b4>] (ctrl_cdev_ioctl+0xd0/0x17c)
[<c02b68b4>] (ctrl_cdev_ioctl) from [<c00bfac4>] (vfs_ioctl+0x18/0x34)
[<c00bfac4>] (vfs_ioctl) from [<c00c059c>] (do_vfs_ioctl+0x530/0x578)
[<c00c059c>] (do_vfs_ioctl) from [<c00c061c>] (SyS_ioctl+0x38/0x54)
[<c00c061c>] (SyS_ioctl) from [<c000db80>] (ret_fast_syscall+0x0/0x30)
UBI error: check_corruption: PEB 354 contains corrupted VID header, and the data does not contain all 0xFF
UBI error: check_corruption: this may be a non-UBI PEB or a severe VID header corruption which requires manual inspection
Volume identifier header dump:
        magic     55424921
        version   1
        vol_type  1
        copy_flag 0
        compat    0
        vol_id    0
        lnum      1
        data_size 0
        used_ebs  0
        data_pad  0
        sqnum     258171
        hdr_crc   f57b76e2
Volume identifier header hexdump:
hexdump of PEB 354 offset 8192, length 1040384
UBI: scanning is finished
UBI error: late_analysis: 1 PEBs are corrupted and preserved
Corrupted PEBs are: 354
UBI assert failed in ubi_wl_init at 1967 (pid 1303)
CPU: 1 PID: 1303 Comm: ubiattach Not tainted 3.14.0-xilinx-00017-g71d29dc-dirty #65
[<c00145c0>] (unwind_backtrace) from [<c0010c50>] (show_stack+0x10/0x14)
[<c0010c50>] (show_stack) from [<c0478df4>] (dump_stack+0x84/0xd4)
[<c0478df4>] (dump_stack) from [<c02be3c8>] (ubi_wl_init+0x3a0/0x460)
[<c02be3c8>] (ubi_wl_init) from [<c02c01b8>] (ubi_attach+0x1154/0x1200)
[<c02c01b8>] (ubi_attach) from [<c02b60a0>] (ubi_attach_mtd_dev+0x59c/0xb60)
[<c02b60a0>] (ubi_attach_mtd_dev) from [<c02b68b4>] (ctrl_cdev_ioctl+0xd0/0x17c)
[<c02b68b4>] (ctrl_cdev_ioctl) from [<c00bfac4>] (vfs_ioctl+0x18/0x34)
[<c00bfac4>] (vfs_ioctl) from [<c00c059c>] (do_vfs_ioctl+0x530/0x578)
[<c00c059c>] (do_vfs_ioctl) from [<c00c061c>] (SyS_ioctl+0x38/0x54)
[<c00c061c>] (SyS_ioctl) from [<c000db80>] (ret_fast_syscall+0x0/0x30)
UBI warning: print_rsvd_warning: cannot reserve enough PEBs for bad PEB handling, reserved 76, need 77
UBI warning: print_rsvd_warning: 1 PEBs are corrupted and not used
UBI: attached mtd2 (name "test2", size 1024 MiB) to ubi0
UBI: PEB size: 1048576 bytes (1024 KiB), LEB size: 1040384 bytes
UBI: min./max. I/O unit sizes: 4096/4096, sub-page size 4096
UBI: VID header offset: 4096 (aligned 4096), data offset: 8192
UBI: good PEBs: 1021, bad PEBs: 3, corrupted PEBs: 1
UBI: user volume: 1, internal volumes: 2, max. volumes count: 128
UBI: max/mean erase counter: 446/254, WL threshold: 4096, image sequence number: 3046330715
UBI: available PEBs: 0, total reserved PEBs: 1020, PEBs reserved for bad PEB handling: 76
UBI: background thread "ubi_bgt0d" started, PID 1306
UBI device number 0, total 1021 LEBs (1062232064 bytes, 1013.0 MiB), available 0 LEBs (0 bytes), LEB size 1040384 bytes (1016.0 KiB)
 Ubi device attach................................[OK] 
 Scan ubi volume..................................[OK] 
 Mount ubifs on tesUBIFS: background thread "ubifs_bgt0_0" started, PID 1311
t dir...............................UBIFS error (pid 1309): ubifs_recover_master_node: failed to recover master node
UBIFS error (pid 1309): ubifs_recover_master_node: dumping second master node
        magic          0x6101831
        crc            0x6a102f75
        node_type      7 (master node)
        group_type     0 (no node group)
        sqnum          5889233
        len            512
        highest_inum   103440
        commit number  247070
        flags          0x3
        log_lnum       4
        root_lnum      937
        root_offs      411568
        root_len       68
        gc_lnum        12
        ihead_lnum     937
        ihead_offs     413696
        index_size     230688
        lpt_lnum       6
        lpt_offs       581727
        nhead_lnum     6
        nhead_offs     585728
        ltab_lnum      6
        ltab_offs      581632
        lsave_lnum     0
        lsave_offs     0
        lscan_lnum     51
        leb_cnt        939
        empty_lebs     880
        idx_lebs       11
        total_free     917319680
        total_dirty    27678232
        total_used     21288136
        total_dead     0
        total_dark     7520256
UBIFS: background thread "ubifs_bgt0_0" stops

Re: mount_ubifs: failed to mount ubifs

[Richard Weinberger]

On Mon, Jun 29, 2015 at 7:03 AM, Bean Huo 霍斌斌 (beanhuo)
<beanhuo@micron.com> wrote:
> Hello:
> Currently ,I mainly focus on ubi powerloss solution for MLC nand ,now I
> Have a problem about master node, below is log.
>
> When PEB 354:4096 read failed,and ubi will regard this block a corrupted block.
> Ubi will not recover it, but reserve it.
> Maybe this block contains master node, so will mount failure?

UBI does not expect that a block turns bad/unreadable all of a sudden.

> Why ubi not try to recover such block?

How would it? If there is an ECC error UBI cannot do much.

The more interesting question is, why went it unreadable?
You said this is MLC nand, UBI was designed for SLC.
MLC is still unsupported, especially if you need to deal with powercuts.

As you work for Micron, it would be wonderful if you could give us more details
on the MLC nand and what Micron recommends to deal with that.

-- 
Thanks,
//richard

RE: mount_ubifs: failed to mount ubifs

[Bean Huo 霍斌斌 (beanhuo)]

>> Hello:
>> Currently ,I mainly focus on ubi powerloss solution for MLC nand ,now 
>> I Have a problem about master node, below is log.
>>
>> When PEB 354:4096 read failed,and ubi will regard this block a corrupted block.
>> Ubi will not recover it, but reserve it.
>> Maybe this block contains master node, so will mount failure?

>UBI does not expect that a block turns bad/unreadable all of a sudden.

>> Why ubi not try to recover such block?

>How would it? If there is an ECC error UBI cannot do much.
This is a question,so need us to do it.

>The more interesting question is, why went it unreadable?
Maybe this page be damaged by a sudden powercut.
>You said this is MLC nand, UBI was designed for SLC.
>MLC is still unsupported, especially if you need to deal with powercuts.

Yes, I am just doing a solution about UBI powerloss for MLC nand.

>As you work for Micron, it would be wonderful if you could give us more details on the MLC nand and what Micron recommends to deal with that.

So far, I am also little know about ubifs. But for MLC nand,maybe I can give you some helpful infor.

>--
>Thanks,
>//richard

Re: mount_ubifs: failed to mount ubifs

[Richard Weinberger]

On Mon, Jun 29, 2015 at 9:03 AM, Bean Huo 霍斌斌 (beanhuo)
<beanhuo@micron.com> wrote:
>>As you work for Micron, it would be wonderful if you could give us more details on the MLC nand and what Micron recommends to deal with that.
>
> So far, I am also little know about ubifs. But for MLC nand,maybe I can give you some helpful infor.

The hardest problem we are aware of are paired pages.
Please search the archives, we had already some discussion on it.

-- 
Thanks,
//richard