1INNOCHECKSUM(1)              MySQL Database System             INNOCHECKSUM(1)
2
3
4

NAME

6       innochecksum - offline InnoDB file checksum utility
7

SYNOPSIS

9       innochecksum [options] file_name
10

DESCRIPTION

12       innochecksum prints checksums for InnoDB files. This tool reads an
13       InnoDB tablespace file, calculates the checksum for each page, compares
14       the calculated checksum to the stored checksum, and reports mismatches,
15       which indicate damaged pages. It was originally developed to speed up
16       verifying the integrity of tablespace files after power outages but can
17       also be used after file copies. Because checksum mismatches cause
18       InnoDB to deliberately shut down a running server, it may be preferable
19       to use this tool rather than waiting for an in-production server to
20       encounter the damaged pages.
21
22       innochecksum cannot be used on tablespace files that the server already
23       has open. For such files, you should use CHECK TABLE to check tables
24       within the tablespace. Attempting to run innochecksum on a tablespace
25       that the server already has open results in an Unable to lock file
26       error.
27
28       If checksum mismatches are found, restore the tablespace from backup or
29       start the server and attempt to use mysqldump to make a backup of the
30       tables within the tablespace.
31
32       Invoke innochecksum like this:
33
34           innochecksum [options] file_name
35
36       innochecksum Options
37
38       innochecksum supports the following options. For options that refer to
39       page numbers, the numbers are zero-based.
40
41--help, -?  Displays command line help. Example usage:
42
43               innochecksum --help
44
45--info, -I Synonym for --help. Displays command line help. Example
46           usage:
47
48               innochecksum --info
49
50--version, -V Displays version information. Example usage:
51
52               innochecksum --version
53
54--verbose, -v Verbose mode; prints a progress indicator to the log
55           file every five seconds. In order for the progress indicator to be
56           printed, the log file must be specified using the --log option. To
57           turn on verbose mode, run:
58
59               innochecksum --verbose
60
61           To turn off verbose mode, run:
62
63               innochecksum --verbose=FALSE
64
65           The --verbose option and --log option can be specified at the same
66           time. For example:
67
68               innochecksum --verbose --log=/var/lib/mysql/test/logtest.txt
69
70           To locate the progress indicator information in the log file, you
71           can perform the following search:
72
73               cat ./logtest.txt | grep -i "okay"
74
75           The progress indicator information in the log file appears similar
76           to the following:
77
78               page 1663 okay: 2.863% done
79               page 8447 okay: 14.537% done
80               page 13695 okay: 23.568% done
81               page 18815 okay: 32.379% done
82               page 23039 okay: 39.648% done
83               page 28351 okay: 48.789% done
84               page 33023 okay: 56.828% done
85               page 37951 okay: 65.308% done
86               page 44095 okay: 75.881% done
87               page 49407 okay: 85.022% done
88               page 54463 okay: 93.722% done
89               ...
90
91--count, -c Print a count of the number of pages in the file and
92           exit. Example usage:
93
94               innochecksum --count ../data/test/tab1.ibd
95
96--start-page=num, -s num Start at this page number. Example usage:
97
98               innochecksum --start-page=600 ../data/test/tab1.ibd
99
100           or:
101
102               innochecksum -s 600 ../data/test/tab1.ibd
103
104--end-page=num, -e num End at this page number. Example usage:
105
106               innochecksum --end-page=700 ../data/test/tab1.ibd
107
108           or:
109
110               innochecksum --p 700 ../data/test/tab1.ibd
111
112--page=num, -p num Check only this page number. Example usage:
113
114               innochecksum --page=701 ../data/test/tab1.ibd
115
116--strict-check, -C Specify a strict checksum algorithm. Options
117           include innodb, crc32, and none.
118
119           In this example, the innodb checksum algorithm is specified:
120
121               innochecksum --strict-check=innodb ../data/test/tab1.ibd
122
123           In this example, the crc32 checksum algorithm is specified:
124
125               innochecksum -C crc32 ../data/test/tab1.ibd
126
127           The following conditions apply:
128
129           •   If you do not specify the --strict-check option, innochecksum
130               validates against innodb, crc32 and none.
131
132           •   If you specify the none option, only checksums generated by
133               none are allowed.
134
135           •   If you specify the innodb option, only checksums generated by
136               innodb are allowed.
137
138           •   If you specify the crc32 option, only checksums generated by
139               crc32 are allowed.
140
141--no-check, -n Ignore the checksum verification when rewriting a
142           checksum. This option may only be used with the innochecksum
143           --write option. If the --write option is not specified,
144           innochecksum terminates.
145
146           In this example, an innodb checksum is rewritten to replace an
147           invalid checksum:
148
149               innochecksum --no-check --write innodb ../data/test/tab1.ibd
150
151--allow-mismatches, -a The maximum number of checksum mismatches
152           allowed before innochecksum terminates. The default setting is 0.
153           If --allow-mismatches=N, where N>=0, N mismatches are permitted and
154           innochecksum terminates at N+1. When --allow-mismatches is set to
155           0, innochecksum terminates on the first checksum mismatch.
156
157           In this example, an existing innodb checksum is rewritten to set
158           --allow-mismatches to 1.
159
160               innochecksum --allow-mismatches=1 --write innodb ../data/test/tab1.ibd
161
162           With --allow-mismatches set to 1, if there is a mismatch at page
163           600 and another at page 700 on a file with 1000 pages, the checksum
164           is updated for pages 0-599 and 601-699. Because --allow-mismatches
165           is set to 1, the checksum tolerates the first mismatch and
166           terminates on the second mismatch, leaving page 600 and pages
167           700-999 unchanged.
168
169--write=name, -w num Rewrite a checksum. When rewriting an invalid
170           checksum, the --no-check option must be used together with the
171           --write option. The --no-check option tells innochecksum to ignore
172           verification of the invalid checksum. You do not have to specify
173           the --no-check option if the current checksum is valid.
174
175           An algorithm must be specified when using the --write option.
176           Possible values for the --write option are:
177
178           •   innodb: A checksum calculated in software, using the original
179               algorithm from InnoDB.
180
181           •   crc32: A checksum calculated using the crc32 algorithm,
182               possibly done with a hardware assist.
183
184           •   none: A constant number.
185
186           The --write option rewrites entire pages to disk. If the new
187           checksum is identical to the existing checksum, the new checksum is
188           not written to disk in order to minimize I/O.
189
190           innochecksum obtains an exclusive lock when the --write option is
191           used.
192
193           In this example, a crc32 checksum is written for tab1.ibd:
194
195               innochecksum -w crc32 ../data/test/tab1.ibd
196
197           In this example, a crc32 checksum is rewritten to replace an
198           invalid crc32 checksum:
199
200               innochecksum --no-check --write crc32 ../data/test/tab1.ibd
201
202--page-type-summary, -S Display a count of each page type in a
203           tablespace. Example usage:
204
205               innochecksum --page-type-summary ../data/test/tab1.ibd
206
207           Sample output for --page-type-summary:
208
209               File::../data/test/tab1.ibd
210               ================PAGE TYPE SUMMARY==============
211               #PAGE_COUNT PAGE_TYPE
212               ===============================================
213                      2        Index page
214                      0        Undo log page
215                      1        Inode page
216                      0        Insert buffer free list page
217                      2        Freshly allocated page
218                      1        Insert buffer bitmap
219                      0        System page
220                      0        Transaction system page
221                      1        File Space Header
222                      0        Extent descriptor page
223                      0        BLOB page
224                      0        Compressed BLOB page
225                      0        Other type of page
226               ===============================================
227               Additional information:
228               Undo page type: 0 insert, 0 update, 0 other
229               Undo page state: 0 active, 0 cached, 0 to_free, 0 to_purge, 0 prepared, 0 other
230
231--page-type-dump, -D Dump the page type information for each page
232           in a tablespace to stderr or stdout. Example usage:
233
234               innochecksum --page-type-dump=/tmp/a.txt ../data/test/tab1.ibd
235
236--log, -l Log output for the innochecksum tool. A log file name
237           must be provided. Log output contains checksum values for each
238           tablespace page. For uncompressed tables, LSN values are also
239           provided. The --log replaces the --debug option, which was
240           available in earlier releases. Example usage:
241
242               innochecksum --log=/tmp/log.txt ../data/test/tab1.ibd
243
244           or:
245
246               innochecksum -l /tmp/log.txt ../data/test/tab1.ibd
247
248       •   - option.  Specify the - option to read from standard input. If the
249           - option is missing when “read from standard in” is expected,
250           innochecksum prints innochecksum usage information indicating that
251           the “-” option was omitted. Example usages:
252
253               cat t1.ibd | innochecksum -
254
255           In this example, innochecksum writes the crc32 checksum algorithm
256           to a.ibd without changing the original t1.ibd file.
257
258               cat t1.ibd | innochecksum --write=crc32 - > a.ibd
259       Running innochecksum on Multiple User-defined Tablespace Files
260
261       The following examples demonstrate how to run innochecksum on multiple
262       user-defined tablespace files (.ibd files).
263
264       Run innochecksum for all tablespace (.ibd) files in the “test”
265       database:
266
267           innochecksum ./data/test/*.ibd
268
269       Run innochecksum for all tablespace files (.ibd files) that have a file
270       name starting with “t”:
271
272           innochecksum ./data/test/t*.ibd
273
274       Run innochecksum for all tablespace files (.ibd files) in the data
275       directory:
276
277           innochecksum ./data/*/*.ibd
278
279
280           Note
281           Running innochecksum on multiple user-defined tablespace files is
282           not supported on Windows operating systems, as Windows shells such
283           as cmd.exe do not support glob pattern expansion. On Windows
284           systems, innochecksum must be run separately for each user-defined
285           tablespace file. For example:
286
287               innochecksum.exe t1.ibd
288               innochecksum.exe t2.ibd
289               innochecksum.exe t3.ibd
290       Running innochecksum on Multiple System Tablespace Files
291
292       By default, there is only one InnoDB system tablespace file (ibdata1)
293       but multiple files for the system tablespace can be defined using the
294       innodb_data_file_path option. In the following example, three files for
295       the system tablespace are defined using the innodb_data_file_path
296       option: ibdata1, ibdata2, and ibdata3.
297
298           ./bin/mysqld --no-defaults --innodb-data-file-path="ibdata1:10M;ibdata2:10M;ibdata3:10M:autoextend"
299
300       The three files (ibdata1, ibdata2, and ibdata3) form one logical system
301       tablespace. To run innochecksum on multiple files that form one logical
302       system tablespace, innochecksum requires the - option to read
303       tablespace files in from standard input, which is equivalent to
304       concatenating multiple files to create one single file. For the example
305       provided above, the following innochecksum command would be used:
306
307           cat ibdata* | innochecksum -
308
309       Refer to the innochecksum options information for more information
310       about the “-” option.
311
312           Note
313           Running innochecksum on multiple files in the same tablespace is
314           not supported on Windows operating systems, as Windows shells such
315           as cmd.exe do not support glob pattern expansion. On Windows
316           systems, innochecksum must be run separately for each system
317           tablespace file. For example:
318
319               innochecksum.exe ibdata1
320               innochecksum.exe ibdata2
321               innochecksum.exe ibdata3
322
324       Copyright © 1997, 2021, Oracle and/or its affiliates.
325
326       This documentation is free software; you can redistribute it and/or
327       modify it only under the terms of the GNU General Public License as
328       published by the Free Software Foundation; version 2 of the License.
329
330       This documentation is distributed in the hope that it will be useful,
331       but WITHOUT ANY WARRANTY; without even the implied warranty of
332       MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
333       General Public License for more details.
334
335       You should have received a copy of the GNU General Public License along
336       with the program; if not, write to the Free Software Foundation, Inc.,
337       51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA or see
338       http://www.gnu.org/licenses/.
339
340

SEE ALSO

342       For more information, please refer to the MySQL Reference Manual, which
343       may already be installed locally and which is also available online at
344       http://dev.mysql.com/doc/.
345

AUTHOR

347       Oracle Corporation (http://dev.mysql.com/).
348
349
350
351MySQL 8.0                         03/07/2021                   INNOCHECKSUM(1)
Impressum