我在 luks 上有一个 ext4 文件系统，基于软件 raid5。几年来，当我开始用完空间时，文件系统一直运行良好。我在 6x2T 驱动器上有一个 9T 卷。我开始升级到 3T 驱动器，方法是执行 mdadm 失败、删除、添加、重建、重复过程，直到我拥有更大的阵列。然后我扩大了 luks 容器，然后当我卸载并尝试 resize2fs 时，我收到消息称文件系统很脏，需要 e2fsck。
我没有多想，就直接执行了 e2fsck -y /dev/mapper/candybox，然后它开始发出各种 inode 被删除类型的消息（记不清具体是什么了），我关闭了 e2fsck，并尝试重新挂载文件系统以备份我所关心的数据。此时尝试挂载时，我得到：
# mount /dev/mapper/candybox /candybox
mount: wrong fs type, bad option, bad superblock on /dev/mapper/candybox,
missing codepage or helper program, or other error
In some cases useful info is found in syslog - try
dmesg | tail or so
回顾我的旧日志，我注意到文件系统每次机器启动时都会出现此错误：
kernel: [79137.275531] EXT4-fs (dm-2): warning: mounting fs with errors, running e2fsck is recommended
我很惭愧自己没注意 :(
然后，我尝试使用每个备份超级块（一个接一个）进行挂载，每次尝试都会在我的日志中留下以下内容：
EXT4-fs (dm-2): ext4_check_descriptors: Checksum for group 0 failed (26534!=65440)
EXT4-fs (dm-2): ext4_check_descriptors: Checksum for group 1 failed (38021!=36729)
EXT4-fs (dm-2): ext4_check_descriptors: Checksum for group 2 failed (18336!=39845)
...
EXT4-fs (dm-2): ext4_check_descriptors: Checksum for group 11911 failed (28743!=44098)
BUG: soft lockup - CPU#0 stuck for 23s! [mount:2939]
尝试重新启动 e2fsck 会导致：
# e2fsck /dev/mapper/candybox
e2fsck 1.41.14 (22-Dec-2010)
e2fsck: Group descriptors look bad... trying backup blocks...
candy: recovering journal
e2fsck: unable to set superblock flags on candy
此时，我决定最好订购更多驱动器并使用ddrescue
现在两周后制作的图像，我在 .img 文件中有一个 luks 分区的图像。
# ls -lh
total 14T
-rw-r--r-- 1 root root 14T Oct 25 01:57 candybox.img
-rw-r--r-- 1 root root 271 Oct 20 14:32 candybox.logfile
尝试了无数次，用尽了所有能在网上找到的方法，我还是无法强制 e2fsck 对镜像做任何事情，所以我使用了，mkfs.ext4 -L candy candybox.img -m 0 -S并且能够在没有日志的情况下以只读方式挂载脏文件系统，并恢复 960G 的数据。它给出了各种错误，比如各种目录不存在等等，但我能够得到一些东西。这给了我一些希望！
然后我再次运行 e2fsck，它必须重新创建根 inode 并给出大量更正组计数列表，我接受了根 inode 创建，并对其他所有问题都说“不”，留下一个完全空的文件系统。再次运行，对所有问题都说“是”，结果相同，但现在是一个“干净”但空的文件系统。
extundelete 给了我0 recoverable inodes found.
现在我又陷入困境了，除了使用 photorec 之类的工具，我实在想不出其他方法，但这会让我陷入文件系统大小的混乱之中。
如果我能得到任何关于如何恢复更多文件的建议或想法，我愿意从原始阵列重新复制图像并重新开始。
我希望能够提供已运行命令的更详细日志，但是除了记录到 syslog 的内容之外，输出内容都被长时间滚动传递，并且由于发生这种情况的时间范围，我的记忆并不那么详细。
任何帮助是极大的赞赏！
10 月 27 日更新
我已经完全重新复制了图像，以便再次开始测试，以下是迄今为止的输出。复制过程：
[root@gamma rescue]# nbd-client 172.16.10.204 2000 /dev/nbd0
Negotiation: ..size = 14307292MB
bs=1024, sz=15002283540480 bytes
[root@gamma rescue]# cryptsetup luksOpen /dev/nbd0 candybox
Enter passphrase for /dev/nbd0:
[root@gamma mnt]# pvcreate /dev/md5
Physical volume "/dev/md5" successfully created
[root@gamma mnt]# pvscan
PV /dev/md5 lvm2 [18.19 TiB]
Total: 1 [18.19 TiB] / in use: 0 [0 ] / in no VG: 1 [18.19 TiB]
[root@gamma mnt]# vgcreate vg-rescue /dev/md5
Volume group "vg-rescue" successfully created
[root@gamma mnt]# lvcreate --size 15T --name lv-rescue vg-rescue
Logical volume "lv-rescue" created
[root@gamma mnt]# mkfs.xfs /dev/vg-rescue/lv-rescue
log stripe unit (524288 bytes) is too large (maximum is 256KiB)
log stripe unit adjusted to 32KiB
meta-data=/dev/vg-rescue/lv-rescue isize=256 agcount=33, agsize=125828992 blks
= sectsz=512 attr=2
data = bsize=4096 blocks=4026531840, imaxpct=5
= sunit=128 swidth=640 blks
naming =version 2 bsize=4096 ascii-ci=0
log =internal log bsize=4096 blocks=521728, version=2
= sectsz=512 sunit=8 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
[root@gamma mnt]# mount /dev/vg-rescue/lv-rescue rescue/
[root@gamma rescue]# ddrescue /dev/mapper/candybox candybox.img candybox.ddlog
Press Ctrl-C to interrupt
Initial status (read from logfile)
rescued: 0 B, errsize: 0 B, errors: 0
Current status
rescued: 13194 GB, errsize: 1807 GB, current rate: 0 B/s
ipos: 13194 GB, errors: 1, average rate: 73528 kB/s
opos: 13194 GB, time from last successful read: 44 s
^Clitting failed blocks...
Interrupted by user
## Network hung, had to try again here
[regan@gamma ~]$ sudo nbd-client -d /dev/nbd0
Disconnecting: que, disconnect, Error: Ioctl failed: Invalid argument
Exiting.
[regan@gamma ~]$ sudo nbd-client 172.16.10.204 2000 /dev/nbd0
Negotiation: ..size = 14307292MB
bs=1024, sz=15002283540480 bytes
[root@gamma rescue]# ddrescue -r 2 /dev/mapper/candybox candybox.img candybox.ddlog
Press Ctrl-C to interrupt
Initial status (read from logfile)
rescued: 15002 GB, errsize: 7426 kB, errors: 60
Current status
rescued: 15002 GB, errsize: 0 B, current rate: 77529 kB/s
ipos: 15002 GB, errors: 0, average rate: 69297 kB/s
opos: 15002 GB, time from last successful read: 0 s
Finished
[root@gamma rescue]# lvcreate -l 100%FREE -s -n rescue_snap /dev/vg-rescue/lv-rescue
Logical volume "rescue_snap" created
[root@gamma rescue]# cd ..
[root@gamma mnt]# mount -o remount,ro rescue/
[root@gamma mnt]# mkdir rescue_snap
[root@gamma mnt]# mount -o nouuid /dev/vg-rescue/rescue_snap rescue_snap
[root@gamma mnt]# cd rescue_snap/
[root@gamma rescue_snap]# ls
candybox.ddlog candybox.img
凌乱的：
[root@gamma rescue_snap]# mkfs.ext4 -L candy candybox.img -m 0 -S
mke2fs 1.41.10 (10-Feb-2009)
candybox.img is not a block special device.
Proceed anyway? (y,n) y
Filesystem label=candy
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
915668992 inodes, 3662666368 blocks
0 blocks (0.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
111776 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
102400000, 214990848, 512000000, 550731776, 644972544, 1934917632,
2560000000
Skipping journal creation in super-only mode
Writing superblocks and filesystem accounting information: done
This filesystem will be automatically checked every 26 mounts or
180 days, whichever comes first. Use tune2fs -c or -i to override.
[root@gamma rescue_snap]# mount -o loop candybox.img /mnt2
[root@gamma rescue_snap]# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/md2 147G 138G 3.1G 98% /
tmpfs 16G 5.7M 16G 1% /dev/shm
/dev/md0 494M 199M 276M 42% /boot
/dev/sdc1 1.8T 979G 763G 57% /mnt/macmirror
/dev/sdj1 1.8T 970G 771G 56% /mnt/usbrescue
/dev/mapper/vg--rescue-lv--rescue
15T 14T 1.4T 91% /mnt/rescue
/dev/mapper/vg--rescue-rescue_snap
15T 14T 1.4T 91% /mnt/rescue_snap
/mnt/rescue_snap/candybox.img
14T 15M 14T 1% /mnt2
## Even though it says only 15M is used, I was able to rsync 960G to /mnt/usbrescue
[root@gamma rescue_snap]# cd /mnt2/
[root@gamma mnt2]# ls -l
ls: cannot access Fedora-19-x86_64-DVD: Input/output error
ls: cannot access rsync_batch: Input/output error
ls: cannot access shell1: Input/output error
ls: cannot access New Folder (2): Input/output error
ls: cannot access shell2: Input/output error
ls: cannot access revolution: Input/output error
ls: cannot access mail: Input/output error
ls: cannot access testing: Input/output error
ls: cannot access export: Input/output error
ls: cannot access ben_backup_20130903: Input/output error
total 160488672
drwxr-xr-x 2 regan regan 4096 Sep 3 20:16 100MEDIA
drwxr-xr-x 19 regan regan 4096 Sep 26 05:18 android
d?????????? ? ? ? ? ? ben_backup_20130903
-rw-rw-r-- 1 regan regan 12126 Jan 4 2013 durations.txt
d?????????? ? ? ? ? ? export
drwxrwxr-x 10 regan regan 4096 Dec 29 2012 family-pc_20121229
d?????????? ? ? ? ? ? Fedora-19-x86_64-DVD
-rw-r--r-- 1 regan regan 72116729363 Sep 30 04:39 gamma_backup_20130928.tgz
-rw-rw-r-- 1 regan regan 55606528323 Jul 27 2011 gamma_tar_20110727.tbz2
-rw-rw-r-- 1 regan regan 3839 Sep 27 2012 Good Quality2.plist
-rw-rw-r-- 1 regan regan 4663 Oct 7 2012 Good Quality3.plist
-rw-rw-r-- 1 regan regan 3852 Sep 26 2012 Good Quality.plist
drwxr-xr-x 7 regan regan 4096 Nov 13 2012 grok
d?????????? ? ? ? ? ? HardDisks
-rwxr--r-- 1 regan regan 54248 Mar 16 2013 IMAG0868.jpg
-rwxr--r-- 1 regan regan 51156 Mar 16 2013 IMAG0869.jpg
-rwxr--r-- 1 regan regan 85912 Mar 16 2013 IMAG0870.jpg
-rwxr--r-- 1 regan regan 76875 Mar 16 2013 IMAG0872.jpg
-rwxr--r-- 1 regan regan 68451 Mar 16 2013 IMAG0873.jpg
-rwxr--r-- 1 regan regan 59587 Mar 16 2013 IMAG0874.jpg
-rwxr--r-- 1 regan regan 81232 Mar 16 2013 IMAG0875.jpg
-rwxr--r-- 1 regan regan 44211 Mar 16 2013 IMAG0876.jpg
-rwxr--r-- 1 regan regan 41660 Mar 16 2013 IMAG0877.jpg
-rwxr--r-- 1 regan regan 36778 Mar 16 2013 IMAG0878.jpg
-rwxr--r-- 1 regan regan 76964 Mar 16 2013 IMAG0879.jpg
-rwxr--r-- 1 regan regan 81876 Mar 16 2013 IMAG0880.jpg
-rwxr--r-- 1 regan regan 1568002 Mar 16 2013 IMAG0953.jpg
-rwxr--r-- 1 regan regan 1548566 Mar 16 2013 IMAG0954.jpg
-rwxr--r-- 1 regan regan 1351743 Mar 16 2013 IMAG0955.jpg
-rwxr--r-- 1 regan regan 1750128 Mar 16 2013 IMAG0956.jpg
-rwxr--r-- 1 regan regan 1694378 Mar 16 2013 IMAG0957.jpg
-rwxr--r-- 1 regan regan 1277128 Mar 16 2013 IMAG0958.jpg
-rwxr--r-- 1 regan regan 1467452 Mar 16 2013 IMAG0965.jpg
-rwxr--r-- 1 regan regan 1595903 Mar 16 2013 IMAG0966.jpg
-rwxr--r-- 1 regan regan 1372444 Mar 16 2013 IMAG0967.jpg
-rwxr--r-- 1 regan regan 1698010 Mar 16 2013 IMAG0968.jpg
-rwxr--r-- 1 regan regan 1550641 Mar 16 2013 IMAG0969.jpg
-rwxr--r-- 1 regan regan 1333768 Mar 16 2013 IMAG0970.jpg
-rwxr--r-- 1 regan regan 1432347 Mar 16 2013 IMAG1010.jpg
-rwxr--r-- 1 regan regan 1668159 Mar 16 2013 IMAG1013.jpg
-rwxr--r-- 1 regan regan 1657058 Mar 16 2013 IMAG1014.jpg
-rwxr--r-- 1 regan regan 1496547 Mar 16 2013 IMAG1016.jpg
-rwxr--r-- 1 regan regan 1609156 Mar 16 2013 IMAG1017.jpg
-rwxr--r-- 1 regan regan 1604832 Mar 16 2013 IMAG1019.jpg
-rwxr--r-- 1 regan regan 2048916 Mar 16 2013 IMAG1073.jpg
-rwxr--r-- 1 regan regan 2006024 Mar 16 2013 IMAG1074.jpg
-rwxr--r-- 1 regan regan 1926686 Mar 16 2013 IMAG1075.jpg
-rw-r--r-- 1 regan regan 1583090 Jul 14 21:15 IMAG1565.jpg
-rw-r--r-- 1 regan regan 1435031 Sep 22 05:19 IMAG1762.jpg
-rw-r--r-- 1 regan regan 1531602 Sep 22 05:19 IMAG1763.jpg
-rw-r--r-- 1 regan regan 1450926 Sep 22 05:19 IMAG1764.jpg
-rw-r--r-- 1 regan regan 1336103 Sep 23 21:31 IMAG1765.jpg
-rw-r--r-- 1 regan regan 1235885 Sep 23 21:32 IMAG1766.jpg
-rw-r--r-- 1 regan regan 1224376 Sep 23 21:32 IMAG1767.jpg
-rw-r--r-- 1 regan regan 1235229 Sep 23 21:32 IMAG1768.jpg
drwxrwxr-x 2 regan regan 4096 Mar 9 2013 jakeanmal
-rw-rw-rw- 1 regan regan 115228 Oct 29 2009 jj_nas
drwx------. 2 root root 16384 Nov 8 2012 lost+found
-rw-r--r-- 1 regan regan 3123877728 Nov 6 2010 luridmirror_20090806.tar.xz
-rw-r--r-- 1 regan regan 2877033943 Mar 1 2013 macabre_20130301.tgz
d?????????? ? ? ? ? ? mail
-rw-r--r-- 1 root root 6771 Aug 10 2009 mail_mirror
-rw------- 1 regan regan 21913047552 Apr 4 2013 mallorys_hdd.vbox.img
d?????????? ? ? ? ? ? MSDN
-rw-r--r-- 1 regan regan 8572 May 10 2010 Music
d?????????? ? ? ? ? ? New Folder (2)
drwxrwxrwx 24 regan regan 4096 Mar 22 2013 onyx
drwxr-xr-x 231 regan regan 24576 Sep 30 09:29 ptp
-rwxr--r-- 1 regan regan 483328 Jan 26 2013 putty.exe
d?????????? ? ? ? ? ? revolution
-rw-r--r-- 1 root root 6272757760 Oct 16 2012 root.tar
d?????????? ? ? ? ? ? rsync_batch
drwxrwxr-x 2 regan regan 12288 Oct 6 04:09 saber
drwxrwxr-x 2 regan regan 188416 Sep 25 04:20 session_tmp
d?????????? ? ? ? ? ? shell1
d?????????? ? ? ? ? ? shell2
d?????????? ? ? ? ? ? testing
drwxrwxr-x 3 regan regan 4096 Oct 7 2012 tofix
-rwxr--r-- 1 regan regan 64991966 Jan 2 2013 VIDEO0041.mp4
[root@gamma mnt2]# cd ..
[root@gamma /]# umount /mnt2
[root@gamma /]# cd /mnt/rescue_snap/
[root@gamma rescue_snap]# e2fsck candybox.img
e2fsck 1.41.10 (10-Feb-2009)
Backing up journal inode block information.
candy contains a file system with errors, check forced.
Resize inode not valid. Recreate<y>? yes
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Root inode not allocated. Allocate<y>? yes
/lost+found not found. Create<y>? yes
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(9252--9255) <Snip a few dozen MB of text> -(3662666237--3662666238) -3662666240 -(3662666242--3662666244) -(3662666247--3662666249) -3662666253 -(3662666255--3662666256) -3662666259 -3662666262 -3662666264 -(3662666268--3662666271) -3662666276 -3662666281 -3662666285 -3662666294 -(3662666296--3662666297) -3662666301 -3662666307 -3662666309 -3662666311 -3662666313 -3662666316 -(3662666318--3662666319) -3662666324 -(3662666326--3662666328) -(3662666331--3662666332) -3662666334 -(3662666341--3662666342) -3662666344 -(3662666346--3662666347) -3662666349 -(3662666351--3662666352) -3662666354 -3662666357 -3662666362 -(3662666366--3662666367)
Fix<y>? yes
Free blocks count wrong for group #0 (23517, counted=23516).
Fix<y>?
Free blocks count wrong (3605188902, counted=3605188901).
Fix<y>? yes
Free inodes count wrong for group #0 (8191, counted=8190).
Fix<y>? yes
Directories count wrong for group #0 (1, counted=2).
Fix<y>? yes
Free inodes count wrong (915668991, counted=915668990).
Fix<y>? yes
candy: ***** FILE SYSTEM WAS MODIFIED *****
candy: 2/915668992 files (0.0% non-contiguous), 57477467/3662666368 blocks
[root@gamma rescue_snap]# mount -o loop candybox.img /mnt2
[root@gamma rescue_snap]# ls -l /mnt2
total 4
drwx------ 2 root root 4096 Oct 27 19:33 lost+found
[root@gamma rescue_snap]#
请注意，我现在在快照中有了备份图像，因此如果有人有想法，我可以反复尝试理论......
答案1
我首选的文件系统数据恢复工具是UFS 探索者。
我不知道 LUKS 加密会对此产生怎样的影响，但请下载 UFS Explorer 评估版并尝试常规最佳实践。针对您的图像运行它，看看是否更容易恢复。该工具将为您提供目录结构的良好视图，并提供一些搜索数据的功能。