It is best to make a backup of a table before performing a table repair operation; under some circumstances the operation might cause data loss. Possible causes include but are not limited to file system errors.

myisamchk [options] tbl_name ...

myisamchk *.MYI

myisamchk /path/to/database_dir/*.MYI

myisamchk /path/to/datadir/*/*.MYI

myisamchk --silent --fast /path/to/datadir/*/*.MYI

myisamchk --silent --force --fast --update-state \


          --key_buffer_size=64M --myisam_sort_buffer_size=64M \


          --read_buffer_size=1M --write_buffer_size=1M \


          /path/to/datadir/*/*.MYI

You must ensure that no other program is using the tables while you are running myisamchk. The most effective means of doing so is to shut down the MySQL server while running myisamchk, or to lock all tables that myisamchk is being used on.

Otherwise, when you run myisamchk, it may display the following error message:

warning: clients are using or haven't closed the table properly

This means that you are trying to check a table that has been updated by another program (such as the mysqld server) that hasn't yet closed the file or that has died without closing the file properly, which can sometimes lead to the corruption of one or more MyISAM tables.

If mysqld is running, you must force it to flush any table modifications that are still buffered in memory by using FLUSH TABLES. You should then ensure that no one is using the tables while you are running myisamchk

However, the easiest way to avoid this problem is to use CHECK TABLE instead of myisamchk to check tables. See Section 13.7.3.2, “CHECK TABLE Statement”.

Display a help message and exit. Options are grouped by type of operation.

Display a help message and exit. Options are presented in a single list.

Write a debugging log. A typical debug_options string is d:t:o,file_name. The default is d:t:o,/tmp/myisamchk.trace.

This option is available only if MySQL was built using WITH_DEBUG. MySQL release binaries provided by Oracle are not built using this option.

Read this option file after the global option file but (on Unix) before the user option file. If the file does not exist or is otherwise inaccessible, an error occurs. If file_name is not an absolute path name, it is interpreted relative to the current directory.

For additional information about this and other option-file options, see Section 4.2.2.3, “Command-Line Options that Affect Option-File Handling”.

Use only the given option file. If the file does not exist or is otherwise inaccessible, an error occurs. If file_name is not an absolute path name, it is interpreted relative to the current directory.

For additional information about this and other option-file options, see Section 4.2.2.3, “Command-Line Options that Affect Option-File Handling”.

Read not only the usual option groups, but also groups with the usual names and a suffix of str. For example, myisamchk normally reads the [myisamchk] group. If this option is given as --defaults-group-suffix=_other, myisamchk also reads the [myisamchk_other] group.

For additional information about this and other option-file options, see Section 4.2.2.3, “Command-Line Options that Affect Option-File Handling”.

Do not read any option files. If program startup fails due to reading unknown options from an option file, --no-defaults can be used to prevent them from being read.

The exception is that the .mylogin.cnf file is read in all cases, if it exists. This permits passwords to be specified in a safer way than on the command line even when --no-defaults is used. To create .mylogin.cnf, use the mysql_config_editor utility. See mysql_config_editor(1).

For additional information about this and other option-file options, see Section 4.2.2.3, “Command-Line Options that Affect Option-File Handling”.

Print the program name and all options that it gets from option files.

For additional information about this and other option-file options, see Section 4.2.2.3, “Command-Line Options that Affect Option-File Handling”.

Silent mode. Write output only when errors occur. You can use -s twice (-ss) to make myisamchk very silent.

Verbose mode. Print more information about what the program does. This can be used with -d and -e. Use -v multiple times (-vv, -vvv) for even more output.

Display version information and exit.

Instead of terminating with an error if the table is locked, wait until the table is unlocked before continuing. If you are running mysqld with external locking disabled, the table can be locked only by another myisamchk command.

myisamchk --recover --ft_min_word_len=3 tbl_name.MYI

[mysqld]
ft_min_word_len=3
[myisamchk]
ft_min_word_len=3

Check the table for errors. This is the default operation if you specify no option that selects an operation type explicitly.

Check only tables that have changed since the last check.

Check the table very thoroughly. This is quite slow if the table has many indexes. This option should only be used in extreme cases. Normally, myisamchk or myisamchk --medium-check should be able to determine whether there are any errors in the table.

If you are using --extend-check and have plenty of memory, setting the key_buffer_size variable to a large value helps the repair operation run faster.

See also the description of this option under table repair options.

For a description of the output format, see the section called “OBTAINING TABLE INFORMATION WITH MYISAMCHK”.

Check only tables that haven't been closed properly.

Do a repair operation automatically if myisamchk finds any errors in the table. The repair type is the same as that specified with the --recover or -r option.

Print informational statistics about the table that is checked.

Do a check that is faster than an --extend-check operation. This finds only 99.99% of all errors, which should be good enough in most cases.

Do not mark the table as checked. This is useful if you use myisamchk to check a table that is in use by some other application that does not use locking, such as mysqld when run with external locking disabled.

Store information in the .MYI file to indicate when the table was checked and whether the table crashed. This should be used to get full benefit of the --check-only-changed option, but you shouldn't use this option if the mysqld server is using the table and you are running it with external locking disabled.

Make a backup of the .MYD file as file_name-time.BAK

The directory where character sets are installed. See Section 10.15, “Character Set Configuration”.

Correct the checksum information for the table.

The maximum length of the data file (when re-creating data file when it is “full”).

Do a repair that tries to recover every possible row from the data file. Normally, this also finds a lot of garbage rows. Do not use this option unless you are desperate.

See also the description of this option under table checking options.

For a description of the output format, see the section called “OBTAINING TABLE INFORMATION WITH MYISAMCHK”.

Overwrite old intermediate files (files with names like tbl_name.TMD) instead of aborting.

For myisamchk, the option value is a bit value that indicates which indexes to update. Each binary bit of the option value corresponds to a table index, where the first index is bit 0. An option value of 0 disables updates to all indexes, which can be used to get faster inserts. Deactivated indexes can be reactivated by using myisamchk -r.

Skip rows larger than the given length if myisamchk cannot allocate memory to hold them.

Achieve a faster repair by modifying only the index file, not the data file. You can specify this option twice to force myisamchk to modify the original data file in case of duplicate keys.

Do a repair that can fix almost any problem except unique keys that are not unique (which is an extremely unlikely error with MyISAM tables). If you want to recover a table, this is the option to try first. You should try --safe-recover only if myisamchk reports that the table cannot be recovered using --recover. (In the unlikely case that --recover fails, the data file remains intact.)

If you have lots of memory, you should increase the value of myisam_sort_buffer_size.

Do a repair using an old recovery method that reads through all rows in order and updates all index trees based on the rows found. This is an order of magnitude slower than --recover, but can handle a couple of very unlikely cases that --recover cannot. This recovery method also uses much less disk space than --recover. Normally, you should repair first using --recover, and then with --safe-recover only if --recover fails.

If you have lots of memory, you should increase the value of key_buffer_size.

Specify the collation to use for sorting table indexes. The character set name is implied by the first part of the collation name.

Force myisamchk to use sorting to resolve the keys even if the temporary files would be very large.

The path of the directory to be used for storing temporary files. If this is not set, myisamchk uses the value of the TMPDIR environment variable. --tmpdir can be set to a list of directory paths that are used successively in round-robin fashion for creating temporary files. The separator character between directory names is the colon (:) on Unix and the semicolon (;) on Windows.

Unpack a table that was packed with myisampack.

Analyze the distribution of key values. This improves join performance by enabling the join optimizer to better choose the order in which to join the tables and which indexes it should use. To obtain information about the key distribution, use a myisamchk --description --verbose tbl_name command or the SHOW INDEX FROM tbl_name statement.

Find the record that a block at the given offset belongs to.

Print some descriptive information about the table. Specifying the --verbose option once or twice produces additional information. See the section called “OBTAINING TABLE INFORMATION WITH MYISAMCHK”.

Sort the index tree blocks in high-low order. This optimizes seeks and makes table scans that use indexes faster.

Sort records according to a particular index. This makes your data much more localized and may speed up range-based SELECT and ORDER BY operations that use this index. (The first time you use this option to sort a table, it may be very slow.) To determine a table's index numbers, use SHOW INDEX, which displays a table's indexes in the same order that myisamchk sees them. Indexes are numbered beginning with 1.

If keys are not packed (PACK_KEYS=0), they have the same length, so when myisamchk sorts and moves records, it just overwrites record offsets in the index. If keys are packed (PACK_KEYS=1), myisamchk must unpack key blocks first, then re-create indexes and pack the key blocks again. (In this case, re-creating indexes is faster than updating offsets for each index.)

Runs myisamchk in “describe mode” to produce a description of your table. If you start the MySQL server with external locking disabled, myisamchk may report an error for a table that is updated while it runs. However, because myisamchk does not change the table in describe mode, there is no risk of destroying data.

Adding -v runs myisamchk in verbose mode so that it produces more information about the table. Adding -v a second time produces even more information.

Shows only the most important information from a table. This operation is slow because it must read the entire table.

This is like -eis, but tells you what is being done.

CREATE TABLE person
(


  id         INT NOT NULL AUTO_INCREMENT,


  last_name  VARCHAR(20) NOT NULL,


  first_name VARCHAR(20) NOT NULL,


  birth      DATE,


  death      DATE,


  PRIMARY KEY (id),


  INDEX (last_name, first_name),


  INDEX (birth)
) MAX_ROWS = 1000000 ENGINE=MYISAM;

-rw-rw----  1 mysql  mysql  9347072 Aug 19 11:47 person.MYD
-rw-rw----  1 mysql  mysql  6066176 Aug 19 11:47 person.MYI

MyISAM file:         person
Record format:       Packed
Character set:       utf8mb4_0900_ai_ci (255)
File-version:        1
Creation time:       2017-03-30 21:21:30
Status:              checked,analyzed,optimized keys,sorted index pages
Auto increment key:              1  Last value:                306688
Data records:               306688  Deleted blocks:                 0
Datafile parts:             306688  Deleted data:                   0
Datafile pointer (bytes):        4  Keyfile pointer (bytes):        3
Datafile length:           9347072  Keyfile length:           6066176
Max datafile length:    4294967294  Max keyfile length:   17179868159
Recordlength:                   54
table description:
Key Start Len Index   Type                     Rec/key         Root  Blocksize
1   2     4   unique  long                           1                    1024
2   6     80  multip. varchar prefix                 0                    1024


    87    80          varchar                        0
3   168   3   multip. uint24 NULL                    0                    1024
Field Start Length Nullpos Nullbit Type
1     1     1
2     2     4                      no zeros
3     6     81                     varchar
4     87    81                     varchar
5     168   3      1       1       no zeros
6     171   3      1       2       no zeros

Name of the MyISAM (index) file.

The format used to store table rows. The preceding examples use Fixed length. Other possible values are Compressed and Packed. (Packed corresponds to what SHOW TABLE STATUS reports as Dynamic.)

The table default character set.

Version of MyISAM format. Always 1.

When the data file was created.

When the index/data file was last reconstructed.

Table status flags. Possible values are crashed, open, changed, analyzed, optimized keys, and sorted index pages.

The key number associated the table's AUTO_INCREMENT column, and the most recently generated value for this column. These fields do not appear if there is no such column.

The number of rows in the table.

How many deleted blocks still have reserved space. You can optimize your table to minimize this space. See Section 7.6.4, “MyISAM Table Optimization”.

For dynamic-row format, this indicates how many data blocks there are. For an optimized table without fragmented rows, this is the same as Data records.

How many bytes of unreclaimed deleted data there are. You can optimize your table to minimize this space. See Section 7.6.4, “MyISAM Table Optimization”.

The size of the data file pointer, in bytes. It is usually 2, 3, 4, or 5 bytes. Most tables manage with 2 bytes, but this cannot be controlled from MySQL yet. For fixed tables, this is a row address. For dynamic tables, this is a byte address.

The size of the index file pointer, in bytes. It is usually 1, 2, or 3 bytes. Most tables manage with 2 bytes, but this is calculated automatically by MySQL. It is always a block address.

How long the table data file can become, in bytes.

How long the table index file can become, in bytes.

How much space each row takes, in bytes.

This key's number. This value is shown only for the first column of the key. If this value is missing, the line corresponds to the second or later column of a multiple-column key. For the table shown in the example, there are two table description lines for the second index. This indicates that it is a multiple-part index with two parts.

Where in the row this portion of the index starts.

How long this portion of the index is. For packed numbers, this should always be the full length of the column. For strings, it may be shorter than the full length of the indexed column, because you can index a prefix of a string column. The total length of a multiple-part key is the sum of the Len values for all key parts.

Whether a key value can exist multiple times in the index. Possible values are unique or multip. (multiple).

What data type this portion of the index has. This is a MyISAM data type with the possible values packed, stripped, or empty.

Address of the root index block.

The size of each index block. By default this is 1024, but the value may be changed at compile time when MySQL is built from source.

This is a statistical value used by the optimizer. It tells how many rows there are per value for this index. A unique index always has a value of 1. This may be updated after a table is loaded (or greatly changed) with myisamchk -a. If this is not updated at all, a default value of 30 is given.

The column number.

The byte position of the column within table rows.

The length of the column in bytes.

For columns that can be NULL, MyISAM stores NULL values as a flag in a byte. Depending on how many nullable columns there are, there can be one or more bytes used for this purpose. The Nullpos and Nullbit values, if nonempty, indicate which byte and bit contains that flag indicating whether the column is NULL.

The position and number of bytes used to store NULL flags is shown in the line for field 1. This is why there are six Field lines for the person table even though it has only five columns.

The data type. The value may contain any of the following descriptors:

The number of the Huffman tree associated with the column.

The number of bits used in the Huffman tree.

Checking MyISAM file: person
Data records:  306688   Deleted blocks:       0
- check file-size
- check record delete-chain
No recordlinks
- check key delete-chain
block_size 1024:
- check index reference
- check data record references index: 1
Key:  1:  Keyblocks used:  98%  Packed:    0%  Max levels:  3
- check data record references index: 2
Key:  2:  Keyblocks used:  99%  Packed:   97%  Max levels:  3
- check data record references index: 3
Key:  3:  Keyblocks used:  98%  Packed:  -14%  Max levels:  3
Total:    Keyblocks used:  98%  Packed:   89%
- check records and index references
*** LOTS OF ROW NUMBERS DELETED ***
Records:            306688  M.recordlength:       25  Packed:            83%
Recordspace used:       97% Empty space:           2% Blocks/Record:   1.00
Record blocks:      306688  Delete blocks:         0
Record data:       7934464  Deleted data:          0
Lost space:         256512  Linkdata:        1156096
User time 43.08, System time 1.68
Maximum resident set size 0, Integral resident set size 0
Non-physical pagefaults 0, Physical pagefaults 0, Swaps 0
Blocks in 0 out 7, Messages in 0 out 0, Signals 0
Voluntary context switches 0, Involuntary context switches 0
Maximum memory usage: 1046926 bytes (1023k)

The number of rows in the table.

How many deleted blocks still have reserved space. You can optimize your table to minimize this space. See Section 7.6.4, “MyISAM Table Optimization”.

The key number.

What percentage of the keyblocks are used. When a table has just been reorganized with myisamchk, the values are very high (very near theoretical maximum).

MySQL tries to pack key values that have a common suffix. This can only be used for indexes on CHAR and VARCHAR columns. For long indexed strings that have similar leftmost parts, this can significantly reduce the space used. In the preceding example, the second key is 40 bytes long and a 97% reduction in space is achieved.

How deep the B-tree for this key is. Large tables with long key values get high values.

How many rows are in the table.

The average row length. This is the exact row length for tables with fixed-length rows, because all rows have the same length.

MySQL strips spaces from the end of strings. The Packed value indicates the percentage of savings achieved by doing this.

What percentage of the data file is used.

What percentage of the data file is unused.

Average number of blocks per row (that is, how many links a fragmented row is composed of). This is always 1.0 for fixed-format tables. This value should stay as close to 1.0 as possible. If it gets too large, you can reorganize the table. See Section 7.6.4, “MyISAM Table Optimization”.

How many blocks (links) are used. For fixed-format tables, this is the same as the number of rows.

How many blocks (links) are deleted.

How many bytes in the data file are used.

How many bytes in the data file are deleted (unused).

If a row is updated to a shorter length, some space is lost. This is the sum of all such losses, in bytes.

When the dynamic table format is used, row fragments are linked with pointers (4 to 7 bytes each). Linkdata is the sum of the amount of storage used by all such pointers.

myisamchk --myisam_sort_buffer_size=256M \


           --key_buffer_size=512M \


           --read_buffer_size=64M \


           --write_buffer_size=64M ...

(largest_key + row_pointer_length) * number_of_rows * 2

You can check the length of the keys and the row_pointer_length with myisamchk -dv tbl_name (see the section called “OBTAINING TABLE INFORMATION WITH MYISAMCHK”). The row_pointer_length and number_of_rows values are the Datafile pointer and Data records values in the table description. To determine the largest_key value, check the Key lines in the table description. The Len column indicates the number of bytes for each key part. For a multiple-column index, the key size is the sum of the Len values for all key parts.