1）用cp命令将SD卡数据拷到Linux下，出现如下错误：

cp: reading `lcn.bin': Input/output error
cp: cannot access `lost+found': Input/output error
cp: reading `navi.bin': Input/output error

出现IO错误，SD卡自动umount，拷贝失败

2）原因可能是SD在不同的文件系统上修改过，导致数据出错。linux下的ext3文件系统不能识别某些文件，拷贝时IO错误

3）用e2fsck /dev/sdb1命令检测修复文件系统，man一下e2fsck

man e2fsck

NAME
       e2fsck - check a Linux ext2/ext3/ext4 file system

SYNOPSIS
       e2fsck  [ -pacnyrdfkvtDFV ] [ -b superblock ] [ -B blocksize ] [ -l|-L bad_blocks_file ] [
       -C fd ] [ -j external-journal ] [ -E extended_options ] device

DESCRIPTION
       e2fsck is used to check the ext2/ext3/ext4 family of file  systems.   For  ext3  and  ext4
       filesystems  that  use  a  journal, if the system has been shut down uncleanly without any
       errors, normally, after replaying the committed transactions  in  the  journal,  the  file
       system  should  be  marked as clean.   Hence, for filesystems that use journalling, e2fsck
       will normally replay the journal and exit, unless its superblock  indicates  that  further
       checking is required.

       device is the device file where the filesystem is stored (e.g.  /dev/hdc1).

       Note that in general it is not safe to run e2fsck on mounted filesystems.  The only excep‐
       tion is if the -n option is specified, and -c, -l, or -L options are not specified.   How‐
       ever,  even  if  it  is  safe to do so, the results printed by e2fsck are not valid if the
       filesystem is mounted.   If e2fsck asks whether or not you should check a filesystem which
       is mounted, the only correct answer is ``no''.  Only experts who really know what they are
       doing should consider answering this question in any other way.

OPTIONS
       -a     This option does the same thing as the -p option.  It  is  provided  for  backwards
              compatibility only; it is suggested that people use -p option whenever possible.

       -b superblock
              Instead  of using the normal superblock, use an alternative superblock specified by

              superblock.  This option is normally used when the primary superblock has been cor‐
              rupted.   The  location  of  the backup superblock is dependent on the filesystem's
              blocksize.  For filesystems with 1k blocksizes, a backup superblock can be found at
              block  8193;  for filesystems with 2k blocksizes, at block 16384; and for 4k block‐
              sizes, at block 32768.

              Additional backup superblocks can be determined by using the mke2fs  program  using
              the  -n  option to print out where the superblocks were created.   The -b option to
              mke2fs, which specifies blocksize of the filesystem must be specified in order  for
              the superblock locations that are printed out to be accurate.

              If  an  alternative  superblock is specified and the filesystem is not opened read-
              only, e2fsck will make sure that the primary superblock  is  updated  appropriately
              upon completion of the filesystem check.

       -B blocksize
              Normally, e2fsck will search for the superblock at various different block sizes in
              an attempt to find the appropriate block size.  This search can be fooled  in  some
              cases.  This option forces e2fsck to only try locating the superblock at a particu‐
              lar blocksize.  If the superblock is not found, e2fsck will terminate with a  fatal
              error.

       -c     This option causes e2fsck to use badblocks(8) program to do a read-only scan of the
              device in order to find any bad blocks.  If any bad  blocks  are  found,  they  are
              added  to  the  bad  block  inode to prevent them from being allocated to a file or
              directory.  If this option is specified twice, then the bad block scan will be done
              using a non-destructive read-write test.

       -C fd  This  option  causes  e2fsck  to write completion information to the specified file
              descriptor so that the progress of the filesystem check  can  be  monitored.   This
              option  is  typically  used  by  programs  which  are  running e2fsck.  If the file
              descriptor number is negative, then absolute value of the file descriptor  will  be
              used,  and  the progress information will be suppressed initially.  It can later be
              enabled by sending the e2fsck process a SIGUSR1 signal.   If  the  file  descriptor
              specified  is  0, e2fsck will print a completion bar as it goes about its business.
              This requires that e2fsck is running on a video console or terminal.

       -d     Print debugging output (useless unless you are debugging e2fsck).

       -D     Optimize directories in filesystem.  This option causes e2fsck to try  to  optimize
              all  directories,  either  by  reindexing them if the filesystem supports directory
              indexing,  or by sorting and compressing directories for  smaller  directories,  or
              for filesystems using traditional linear directories.

              Even without the -D option, e2fsck may sometimes optimize a few directories --- for
              example, if directory indexing is enabled and a directory is not indexed and  would
              benefit from being indexed, or if the index structures are corrupted and need to be
              rebuilt.  The -D option forces all directories in the filesystem to  be  optimized.
              This can sometimes make them a little smaller and slightly faster to search, but in
              practice, you should rarely need to use this option.

              The -D option will detect directory entries with duplicate names in a single direc‐
              tory, which e2fsck normally does not enforce for performance reasons.

       -E extended_options
              Set e2fsck extended options.  Extended options are comma separated, and may take an
              argument using the equals ('=') sign.  The following options are supported:

                   ea_ver=extended_attribute_version
                          Set the version of the extended  attribute  blocks  which  e2fsck  will
                          require  while checking the filesystem.  The version number may be 1 or
                          2.  The default extended attribute version format is 2.

                   fragcheck
                          During pass 1, print a detailed report of any discontiguous blocks  for
                          files in the filesystem.

       -f     Force checking even if the file system seems clean.

       -F     Flush  the  filesystem device's buffer caches before beginning.  Only really useful
              for doing e2fsck time trials.

       -j external-journal
              Set the pathname where the external-journal for this filesystem can be found.

       -k     When combined with the -c option, any existing bad blocks in the  bad  blocks  list
              are  preserved,  and any new bad blocks found by running badblocks(8) will be added
              to the existing bad blocks list.

       -l filename
              Add the block numbers listed in the file specified by filename to the list  of  bad
              blocks.   The  format  of  this  file  is the same as the one generated by the bad‐
              blocks(8) program.  Note that the block numbers are based on the blocksize  of  the
              filesystem.   Hence,  badblocks(8) must be given the blocksize of the filesystem in
              order to obtain correct results.  As a result, it is much simpler and safer to  use
              the  -c  option  to  e2fsck,  since  it will assure that the correct parameters are
              passed to the badblocks program.

       -L filename
              Set the bad blocks list to be the list of  blocks  specified  by  filename.   (This
              option  is  the same as the -l option, except the bad blocks list is cleared before
              the blocks listed in the file are added to the bad blocks list.)

       -n     Open the filesystem read-only, and assume an  answer  of  `no'  to  all  questions.
              Allows  e2fsck  to  be used non-interactively.  This option may not be specified at
              the same time as the -p or -y options.

       -p     Automatically repair ("preen") the file system.  This option will cause  e2fsck  to
              automatically  fix  any  filesystem problems that can be safely fixed without human
              intervention.  If e2fsck discovers a problem which may require the system  adminis‐
              trator to take additional corrective action, e2fsck will print a description of the
              problem and then exit with the value 4 logically or'ed into the  exit  code.   (See
              the EXIT CODE section.)  This option is normally used by the system's boot scripts.
              It may not be specified at the same time as the -n or -y options.

       -r     This option does nothing at all; it is provided only for backwards compatibility.

       -t     Print timing statistics for e2fsck.  If this option is used twice, additional  tim‐
              ing statistics are printed on a pass by pass basis.

       -v     Verbose mode.

       -V     Print version information and exit.

       -y     Assume  an  answer of `yes' to all questions; allows e2fsck to be used non-interac‐
              tively.  This option may not be specified at the same time as the -n or -p options.

EXIT CODE
       The exit code returned by e2fsck is the sum of the following conditions:
            0    - No errors
            1    - File system errors corrected
            2    - File system errors corrected, system should
                   be rebooted
            4    - File system errors left uncorrected
            8    - Operational error
            16   - Usage or syntax error
            32   - E2fsck canceled by user request
            128  - Shared library error

SIGNALS
       The following signals have the following effect when sent to e2fsck.

       SIGUSR1
              This signal causes e2fsck to start displaying a completion bar or emitting progress
              information.  (See discussion of the -C option.)

       SIGUSR2
              This  signal causes e2fsck to stop displaying a completion bar or emitting progress
              information.

REPORTING BUGS
       Almost any piece of software will have bugs.  If you manage to  find  a  filesystem  which
       causes  e2fsck  to  crash,  or  which  e2fsck is unable to repair, please report it to the
       author.

       Please include as much information as possible in your bug  report.   Ideally,  include  a
       complete  transcript  of the e2fsck run, so I can see exactly what error messages are dis‐
       played.  (Make sure the messages printed by e2fsck are in English; if your system has been
       configured  so  that  e2fsck's messages have been translated into another language, please
       set the the LC_ALL environment variable to C so that the  transcript  of  e2fsck's  output
       will  be  useful  to  me.)   If you have a writable filesystem where the transcript can be
       stored, the script(1) program is a handy way to save the output of e2fsck to a file.

       It is also useful to send the output of dumpe2fs(8).  If a specific inode or inodes  seems
       to be giving e2fsck trouble, try running the debugfs(8) command and send the output of the
       stat(1u) command run on the relevant inode(s).  If the inode is a directory,  the  debugfs
       dump command will allow you to extract the contents of the directory inode, which can sent
       to me after being first run through uuencode(1).  The most useful data  you  can  send  to
       help  reproduce  the bug is a compressed raw image dump of the filesystem, generated using
       e2image(8).  See the e2image(8) man page for more details.

       Always include the full version string which e2fsck displays when it is  run,  so  I  know
       which version you are running.

AUTHOR
       This version of e2fsck was written by Theodore Ts'o <tytso@mit.edu>.

SEE ALSO
       e2fsck.conf(5), badblocks(8), dumpe2fs(8), debugfs(8), e2image(8), mke2fs(8), tune2fs(8)

4）e2fsck 之后，可以正常拷贝。

2011.01.06