"crc32 mismatch" followed by segfault in btrfs receive

[Alex Jordan <alex@strugee.net>]

[-- Attachment #1: Type: text/plain, Size: 2872 bytes --]

Hey, all -

Recently my btrfs filesystem started hanging forever when mounted
read-write. However, mounting it readonly succeeded, so I backed up my
oldest backup subvolume with `btrfs send` and then saved the rest of
the backup subvolumes with `btrfs send -p`. Then I reformatted the
filesystem.

However, upon trying to restore the initial subvolume with `btrfs
receive -vvvvv --max-errors 10`, I get:

    mkfile o28626690-3011443-0
    rename o28626690-3011443-0 -> var/lib/logrotate/status
    utimes var/lib/logrotate
    truncate var/lib/logrotate/status size=3D18897
    chown var/lib/logrotate/status - uid=3D0, gid=3D0
    chmod var/lib/logrotate/status - mode=3D0644
    utimes var/lib/logrotate/status
    mkfile o28626715-3011443-0
    rename o28626715-3011443-0 -> var/lib/mlocate/mlocate.db
    utimes var/lib/mlocate
    ERROR: crc32 mismatch in command.
    ERROR: read from stream failed. Bad address
    ERROR: crc32 mismatch in command.
    ERROR: read from stream failed. Bad address
    ERROR: read from stream failed. Bad address
    Segmentation fault

at the end. Any idea what could be going on?

Here are some details about my system:

    % uname -a
    Linux steevie 4.7.0-0.bpo.1-amd64 #1 SMP Debian 4.7.8-1~bpo8+1 (2016-10-19) x86_64 GNU/Linux
    
    % btrfs --version # Installed from Debian's jessie-backports
    btrfs-progs v4.7.3
    
    % sudo btrfs fi show
    [sudo] password for alex: 
    Label: '[root]'  uuid: 06c6565d-af6c-4123-9e42-3b6281f164e6
            Total devices 3 FS bytes used 430.48GiB
            devid    1 size 109.45GiB used 1.00GiB path /dev/mapper/sdc6_crypt
            devid    2 size 931.47GiB used 460.04GiB path /dev/mapper/sda2_crypt
            devid    3 size 931.47GiB used 461.04GiB path /dev/mapper/sdb2_crypt
    
    warning, device 4 is missing
    warning, device 3 is missing
    bytenr mismatch, want=895655870464, have=0
    ERROR: cannot read chunk root
    Label: '[root]'  uuid: 1c3126e2-2834-475b-92f2-853c1e13552b
            Total devices 4 FS bytes used 414.79GiB
            devid    5 size 109.45GiB used 3.00GiB path /dev/sdd6
            *** Some devices missing
    
    % sudo btrfs fi df /                               
    Data, RAID1: total=455.00GiB, used=426.58GiB
    System, RAID1: total=40.00MiB, used=96.00KiB
    Metadata, RAID1: total=6.00GiB, used=3.90GiB
    GlobalReserve, single: total=512.00MiB, used=0.00B

dmesg has a single relevant line:

> traps: btrfs[27981] general protection ip:7fec861d81bd sp:7ffff8911240 error:0 in libc-2.19.so[7fec861a1000+1a1000]

The entire log is available at http://sprunge.us/UQWP. I know
pastebin-type services are discouraged on here, so sorry about that -
it's ~112K so my email got rejected when I sent it as an attachment.

Thanks in advance,

AJ

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 963 bytes --]