.\" Automatically generated by Pod::Man 4.11 (Pod::Simple 3.35) .\" .\" Standard preamble: .\" ======================================================================== .de Sp \" Vertical space (when we can't use .PP) .if t .sp .5v .if n .sp .. .de Vb \" Begin verbatim text .ft CW .nf .ne \\$1 .. .de Ve \" End verbatim text .ft R .fi .. .\" Set up some character translations and predefined strings. \*(-- will .\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left .\" double quote, and \*(R" will give a right double quote. \*(C+ will .\" give a nicer C++. Capital omega is used to do unbreakable dashes and .\" therefore won't be available. \*(C` and \*(C' expand to `' in nroff, .\" nothing in troff, for use with C<>. .tr \(*W- .ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p' .ie n \{\ . ds -- \(*W- . ds PI pi . if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch . if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\" diablo 12 pitch . ds L" "" . ds R" "" . ds C` "" . ds C' "" 'br\} .el\{\ . ds -- \|\(em\| . ds PI \(*p . ds L" `` . ds R" '' . ds C` . ds C' 'br\} .\" .\" Escape single quotes in literal strings from groff's Unicode transform. .ie \n(.g .ds Aq \(aq .el .ds Aq ' .\" .\" If the F register is >0, we'll generate index entries on stderr for .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index .\" entries marked with X<> in POD. Of course, you'll have to process the .\" output yourself in some meaningful fashion. .\" .\" Avoid warning from groff about undefined register 'F'. .de IX .. .nr rF 0 .if \n(.g .if rF .nr rF 1 .if (\n(rF:(\n(.g==0)) \{\ . if \nF \{\ . de IX . tm Index:\\$1\t\\n%\t"\\$2" .. . if !\nF==2 \{\ . nr % 0 . nr F 2 . \} . \} .\} .rr rF .\" ======================================================================== .\" .IX Title "PT-TABLE-SYNC 1" .TH PT-TABLE-SYNC 1 "2022-07-07" "perl v5.26.3" "User Contributed Perl Documentation" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l .nh .SH "NAME" pt\-table\-sync \- Synchronize MySQL table data efficiently. .SH "SYNOPSIS" .IX Header "SYNOPSIS" Usage: pt-table-sync [\s-1OPTIONS\s0] \s-1DSN\s0 [\s-1DSN\s0] .PP pt-table-sync synchronizes data efficiently between MySQL tables. .PP This tool changes data, so for maximum safety, you should back up your data before using it. When synchronizing a server that is a replication slave with the \*(L"\-\-replicate\*(R" or \*(L"\-\-sync\-to\-master\*(R" methods, it \fBalways\fR makes the changes on the replication master, \fBnever\fR the replication slave directly. This is in general the only safe way to bring a replica back in sync with its master; changes to the replica are usually the source of the problems in the first place. However, the changes it makes on the master should be no-op changes that set the data to their current values, and actually affect only the replica. .PP Sync db.tbl on host1 to host2: .PP .Vb 1 \& pt\-table\-sync \-\-execute h=host1,D=db,t=tbl h=host2 .Ve .PP Sync all tables on host1 to host2 and host3: .PP .Vb 1 \& pt\-table\-sync \-\-execute host1 host2 host3 .Ve .PP Make slave1 have the same data as its replication master: .PP .Vb 1 \& pt\-table\-sync \-\-execute \-\-sync\-to\-master slave1 .Ve .PP Resolve differences that pt-table-checksum found on all slaves of master1: .PP .Vb 1 \& pt\-table\-sync \-\-execute \-\-replicate test.checksum master1 .Ve .PP Same as above but only resolve differences on slave1: .PP .Vb 2 \& pt\-table\-sync \-\-execute \-\-replicate test.checksum \e \& \-\-sync\-to\-master slave1 .Ve .PP Sync master2 in a master-master replication configuration, where master2's copy of db.tbl is known or suspected to be incorrect: .PP .Vb 1 \& pt\-table\-sync \-\-execute \-\-sync\-to\-master h=master2,D=db,t=tbl .Ve .PP Note that in the master-master configuration, the following will \s-1NOT\s0 do what you want, because it will make changes directly on master2, which will then flow through replication and change master1's data: .PP .Vb 2 \& # Don\*(Aqt do this in a master\-master setup! \& pt\-table\-sync \-\-execute h=master1,D=db,t=tbl master2 .Ve .SH "RISKS" .IX Header "RISKS" \&\fB\s-1WARNING\s0\fR: pt-table-sync changes data! Before using this tool, please: .IP "\(bu" 4 Read the tool's documentation .IP "\(bu" 4 Review the tool's known \*(L"\s-1BUGS\*(R"\s0 .IP "\(bu" 4 Test the tool on a non-production server .IP "\(bu" 4 Backup your production server and verify the backups .Sp pt-table-sync is mature, proven in the real world, and well tested, but if used improperly it can have adverse consequences. Always test syncing first with \*(L"\-\-dry\-run\*(R" and \*(L"\-\-print\*(R". .SH "DESCRIPTION" .IX Header "DESCRIPTION" pt-table-sync does one-way and bidirectional synchronization of table data. It does \fBnot\fR synchronize table structures, indexes, or any other schema objects. The following describes one-way synchronization. \&\*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 is described later. .PP This tool is complex and functions in several different ways. To use it safely and effectively, you should understand three things: the purpose of \*(L"\-\-replicate\*(R", finding differences, and specifying hosts. These three concepts are closely related and determine how the tool will run. The following is the abbreviated logic: .PP .Vb 10 \& if DSN has a t part, sync only that table: \& if 1 DSN: \& if \-\-sync\-to\-master: \& The DSN is a slave. Connect to its master and sync. \& if more than 1 DSN: \& The first DSN is the source. Sync each DSN in turn. \& else if \-\-replicate: \& if \-\-sync\-to\-master: \& The DSN is a slave. Connect to its master, find records \& of differences, and fix. \& else: \& The DSN is the master. Find slaves and connect to each, \& find records of differences, and fix. \& else: \& if only 1 DSN and \-\-sync\-to\-master: \& The DSN is a slave. Connect to its master, find tables and \& filter with \-\-databases etc, and sync each table to the master. \& else: \& find tables, filtering with \-\-databases etc, and sync each \& DSN to the first. .Ve .PP pt-table-sync can run in one of two ways: with \*(L"\-\-replicate\*(R" or without. The default is to run without \*(L"\-\-replicate\*(R" which causes pt-table-sync to automatically find differences efficiently with one of several algorithms (see \*(L"\s-1ALGORITHMS\*(R"\s0). Alternatively, the value of \&\*(L"\-\-replicate\*(R", if specified, causes pt-table-sync to use the differences already found by having previously ran pt-table-checksum with its own \&\f(CW\*(C`\-\-replicate\*(C'\fR option. Strictly speaking, you don't need to use \&\*(L"\-\-replicate\*(R" because pt-table-sync can find differences, but many people use \*(L"\-\-replicate\*(R" if, for example, they checksum regularly using pt-table-checksum then fix differences as needed with pt-table-sync. If you're unsure, read each tool's documentation carefully and decide for yourself, or consult with an expert. .PP Regardless of whether \*(L"\-\-replicate\*(R" is used or not, you need to specify which hosts to sync. There are two ways: with \*(L"\-\-sync\-to\-master\*(R" or without. Specifying \*(L"\-\-sync\-to\-master\*(R" makes pt-table-sync expect one and only slave \s-1DSN\s0 on the command line. The tool will automatically discover the slave's master and sync it so that its data is the same as its master. This is accomplished by making changes on the master which then flow through replication and update the slave to resolve its differences. \&\fBBe careful though\fR: although this option specifies and syncs a single slave, if there are other slaves on the same master, they will receive via replication the changes intended for the slave that you're trying to sync. .PP Alternatively, if you do not specify \*(L"\-\-sync\-to\-master\*(R", the first \&\s-1DSN\s0 given on the command line is the source host. There is only ever one source host. If you do not also specify \*(L"\-\-replicate\*(R", then you must specify at least one other \s-1DSN\s0 as the destination host. There can be one or more destination hosts. Source and destination hosts must be independent; they cannot be in the same replication topology. pt-table-sync will die with an error if it detects that a destination host is a slave because changes are written directly to destination hosts (and it's not safe to write directly to slaves). Or, if you specify \&\*(L"\-\-replicate\*(R" (but not \*(L"\-\-sync\-to\-master\*(R") then pt-table-sync expects one and only one master \s-1DSN\s0 on the command line. The tool will automatically discover all the master's slaves and sync them to the master. This is the only way to sync several (all) slaves at once (because \&\*(L"\-\-sync\-to\-master\*(R" only specifies one slave). .PP Each host on the command line is specified as a \s-1DSN.\s0 The first \s-1DSN\s0 (or only \s-1DSN\s0 for cases like \*(L"\-\-sync\-to\-master\*(R") provides default values for other DSNs, whether those other DSNs are specified on the command line or auto-discovered by the tool. So in this example, .PP .Vb 1 \& pt\-table\-sync \-\-execute h=host1,u=msandbox,p=msandbox h=host2 .Ve .PP the host2 \s-1DSN\s0 inherits the \f(CW\*(C`u\*(C'\fR and \f(CW\*(C`p\*(C'\fR \s-1DSN\s0 parts from the host1 \s-1DSN.\s0 Use the \*(L"\-\-explain\-hosts\*(R" option to see how pt-table-sync will interpret the DSNs given on the command line. .SH "LIMITATIONS" .IX Header "LIMITATIONS" .IP "Replicas using row-based replication" 4 .IX Item "Replicas using row-based replication" pt-table-sync requires statement-based replication when used with the \*(L"\-\-sync\-to\-master\*(R" or \*(L"\-\-replicate\*(R" option. Therefore it will set \f(CW\*(C`binlog_format=STATEMENT\*(C'\fR on the master for its session if required. To do this user must have \f(CW\*(C`SUPER\*(C'\fR privilege. .SH "OUTPUT" .IX Header "OUTPUT" If you specify the \*(L"\-\-verbose\*(R" option, you'll see information about the differences between the tables. There is one row per table. Each server is printed separately. For example, .PP .Vb 3 \& # Syncing h=host1,D=test,t=test1 \& # DELETE REPLACE INSERT UPDATE ALGORITHM START END EXIT DATABASE.TABLE \& # 0 0 3 0 Chunk 13:00:00 13:00:17 2 test.test1 .Ve .PP Table test.test1 on host1 required 3 \f(CW\*(C`INSERT\*(C'\fR statements to synchronize and it used the Chunk algorithm (see \*(L"\s-1ALGORITHMS\*(R"\s0). The sync operation for this table started at 13:00:00 and ended 17 seconds later (times taken from \f(CW\*(C`NOW()\*(C'\fR on the source host). Because differences were found, its \&\*(L"\s-1EXIT STATUS\*(R"\s0 was 2. .PP If you specify the \*(L"\-\-print\*(R" option, you'll see the actual \s-1SQL\s0 statements that the script uses to synchronize the table if \*(L"\-\-execute\*(R" is also specified. .PP If you want to see the \s-1SQL\s0 statements that pt-table-sync is using to select chunks, nibbles, rows, etc., then specify \*(L"\-\-print\*(R" once and \*(L"\-\-verbose\*(R" twice. Be careful though: this can print a lot of \s-1SQL\s0 statements. .PP There are cases where no combination of \f(CW\*(C`INSERT\*(C'\fR, \f(CW\*(C`UPDATE\*(C'\fR or \f(CW\*(C`DELETE\*(C'\fR statements can resolve differences without violating some unique key. For example, suppose there's a primary key on column a and a unique key on column b. Then there is no way to sync these two tables with straightforward \s-1UPDATE\s0 statements: .PP .Vb 6 \& +\-\-\-+\-\-\-+ +\-\-\-+\-\-\-+ \& | a | b | | a | b | \& +\-\-\-+\-\-\-+ +\-\-\-+\-\-\-+ \& | 1 | 2 | | 1 | 1 | \& | 2 | 1 | | 2 | 2 | \& +\-\-\-+\-\-\-+ +\-\-\-+\-\-\-+ .Ve .PP The tool rewrites queries to \f(CW\*(C`DELETE\*(C'\fR and \f(CW\*(C`REPLACE\*(C'\fR in this case. This is automatically handled after the first index violation, so you don't have to worry about it. .PP Be careful when using pt-table-sync in any master-master setup. Master-master replication is inherently tricky, and it's easy to make mistakes. You need to be sure you're using the tool correctly for master-master replication. See the \&\*(L"\s-1SYNOPSIS\*(R"\s0 for the overview of the correct usage. .PP Also be careful with tables that have foreign key constraints with \f(CW\*(C`ON DELETE\*(C'\fR or \f(CW\*(C`ON UPDATE\*(C'\fR definitions because these might cause unintended changes on the child tables. See \*(L"\-\-[no]check\-child\-tables\*(R". .PP In general, this tool is best suited when your tables have a primary key or unique index. Although it can synchronize data in tables lacking a primary key or unique index, it might be best to synchronize that data by another means. .SH "REPLICATION SAFETY" .IX Header "REPLICATION SAFETY" Synchronizing a replication master and slave safely is a non-trivial problem, in general. There are all sorts of issues to think about, such as other processes changing data, trying to change data on the slave, whether the destination and source are a master-master pair, and much more. .PP In general, the safe way to do it is to change the data on the master, and let the changes flow through replication to the slave like any other changes. However, this works only if it's possible to \s-1REPLACE\s0 into the table on the master. \s-1REPLACE\s0 works only if there's a unique index on the table (otherwise it just acts like an ordinary \s-1INSERT\s0). .PP If your table has unique keys, you should use the \*(L"\-\-sync\-to\-master\*(R" and/or \&\*(L"\-\-replicate\*(R" options to sync a slave to its master. This will generally do the right thing. When there is no unique key on the table, there is no choice but to change the data on the slave, and pt-table-sync will detect that you're trying to do so. It will complain and die unless you specify \&\f(CW\*(C`\-\-no\-check\-slave\*(C'\fR (see \*(L"\-\-[no]check\-slave\*(R"). .PP If you're syncing a table without a primary or unique key on a master-master pair, you must change the data on the destination server. Therefore, you need to specify \f(CW\*(C`\-\-no\-bin\-log\*(C'\fR for safety (see \*(L"\-\-[no]bin\-log\*(R"). If you don't, the changes you make on the destination server will replicate back to the source server and change the data there! .PP The generally safe thing to do on a master-master pair is to use the \&\*(L"\-\-sync\-to\-master\*(R" option so you don't change the data on the destination server. You will also need to specify \f(CW\*(C`\-\-no\-check\-slave\*(C'\fR to keep pt-table-sync from complaining that it is changing data on a slave. .SH "ALGORITHMS" .IX Header "ALGORITHMS" pt-table-sync has a generic data-syncing framework which uses different algorithms to find differences. The tool automatically chooses the best algorithm for each table based on indexes, column types, and the algorithm preferences specified by \*(L"\-\-algorithms\*(R". The following algorithms are available, listed in their default order of preference: .IP "Chunk" 4 .IX Item "Chunk" Finds an index whose first column is numeric (including date and time types), and divides the column's range of values into chunks of approximately \&\*(L"\-\-chunk\-size\*(R" rows. Syncs a chunk at a time by checksumming the entire chunk. If the chunk differs on the source and destination, checksums each chunk's rows individually to find the rows that differ. .Sp It is efficient when the column has sufficient cardinality to make the chunks end up about the right size. .Sp The initial per-chunk checksum is quite small and results in minimal network traffic and memory consumption. If a chunk's rows must be examined, only the primary key columns and a checksum are sent over the network, not the entire row. If a row is found to be different, the entire row will be fetched, but not before. .Sp Note that this algorithm will not work if chunking a char column where all the values start with the same character. In that case, the tool will exit and suggest picking a different algorithm. .IP "Nibble" 4 .IX Item "Nibble" Finds an index and ascends the index in fixed-size nibbles of \*(L"\-\-chunk\-size\*(R" rows, using a non-backtracking algorithm (see pt-archiver for more on this algorithm). It is very similar to \*(L"Chunk\*(R", but instead of pre-calculating the boundaries of each piece of the table based on index cardinality, it uses \&\f(CW\*(C`LIMIT\*(C'\fR to define each nibble's upper limit, and the previous nibble's upper limit to define the lower limit. .Sp It works in steps: one query finds the row that will define the next nibble's upper boundary, and the next query checksums the entire nibble. If the nibble differs between the source and destination, it examines the nibble row-by-row, just as \*(L"Chunk\*(R" does. .IP "GroupBy" 4 .IX Item "GroupBy" Selects the entire table grouped by all columns, with a \s-1COUNT\s0(*) column added. Compares all columns, and if they're the same, compares the \s-1COUNT\s0(*) column's value to determine how many rows to insert or delete into the destination. Works on tables with no primary key or unique index. .IP "Stream" 4 .IX Item "Stream" Selects the entire table in one big stream and compares all columns. Selects all columns. Much less efficient than the other algorithms, but works when there is no suitable index for them to use. .IP "Future Plans" 4 .IX Item "Future Plans" Possibilities for future algorithms are TempTable (what I originally called bottom-up in earlier versions of this tool), DrillDown (what I originally called top-down), and GroupByPrefix (similar to how SqlYOG Job Agent works). Each algorithm has strengths and weaknesses. If you'd like to implement your favorite technique for finding differences between two sources of data on possibly different servers, I'm willing to help. The algorithms adhere to a simple interface that makes it pretty easy to write your own. .SH "BIDIRECTIONAL SYNCING" .IX Header "BIDIRECTIONAL SYNCING" Bidirectional syncing is a new, experimental feature. To make it work reliably there are a number of strict limitations: .PP .Vb 5 \& * only works when syncing one server to other independent servers \& * does not work in any way with replication \& * requires that the table(s) are chunkable with the Chunk algorithm \& * is not N\-way, only bidirectional between two servers at a time \& * does not handle DELETE changes .Ve .PP For example, suppose we have three servers: c1, r1, r2. c1 is the central server, a pseudo-master to the other servers (viz. r1 and r2 are not slaves to c1). r1 and r2 are remote servers. Rows in table foo are updated and inserted on all three servers and we want to synchronize all the changes between all the servers. Table foo has columns: .PP .Vb 3 \& id int PRIMARY KEY \& ts timestamp auto updated \& name varchar .Ve .PP Auto-increment offsets are used so that new rows from any server do not create conflicting primary key (id) values. In general, newer rows, as determined by the ts column, take precedence when a same but differing row is found during the bidirectional sync. \*(L"Same but differing\*(R" means that two rows have the same primary key (id) value but different values for some other column, like the name column in this example. Same but differing conflicts are resolved by a \*(L"conflict\*(R". A conflict compares some column of the competing rows to determine a \*(L"winner\*(R". The winning row becomes the source and its values are used to update the other row. .PP There are subtle differences between three columns used to achieve bidirectional syncing that you should be familiar with: chunk column (\*(L"\-\-chunk\-column\*(R"), comparison column(s) (\*(L"\-\-columns\*(R"), and conflict column (\*(L"\-\-conflict\-column\*(R"). The chunk column is only used to chunk the table; e.g. \*(L"\s-1WHERE\s0 id >= 5 \s-1AND\s0 id < 10\*(R". Chunks are checksummed and when chunk checksums reveal a difference, the tool selects the rows in that chunk and checksums the \*(L"\-\-columns\*(R" for each row. If a column checksum differs, the rows have one or more conflicting column values. In a traditional unidirectional sync, the conflict is a moot point because it can be resolved simply by updating the entire destination row with the source row's values. In a bidirectional sync, however, the \*(L"\-\-conflict\-column\*(R" (in accordance with other \f(CW\*(C`\-\-conflict\-*\*(C'\fR options list below) is compared to determine which row is \*(L"correct\*(R" or \*(L"authoritative\*(R"; this row becomes the \*(L"source\*(R". .PP To sync all three servers completely, two runs of pt-table-sync are required. The first run syncs c1 and r1, then syncs c1 and r2 including any changes from r1. At this point c1 and r2 are completely in sync, but r1 is missing any changes from r2 because c1 didn't have these changes when it and r1 were synced. So a second run is needed which syncs the servers in the same order, but this time when c1 and r1 are synced r1 gets r2's changes. .PP The tool does not sync N\-ways, only bidirectionally between the first \s-1DSN\s0 given on the command line and each subsequent \s-1DSN\s0 in turn. So the tool in this example would be ran twice like: .PP .Vb 1 \& pt\-table\-sync \-\-bidirectional h=c1 h=r1 h=r2 .Ve .PP The \*(L"\-\-bidirectional\*(R" option enables this feature and causes various sanity checks to be performed. You must specify other options that tell pt-table-sync how to resolve conflicts for same but differing rows. These options are: .PP .Vb 5 \& * \-\-conflict\-column \& * \-\-conflict\-comparison \& * \-\-conflict\-value \& * \-\-conflict\-threshold \& * \-\-conflict\-error"> (optional) .Ve .PP Use \*(L"\-\-print\*(R" to test this option before \*(L"\-\-execute\*(R". The printed \&\s-1SQL\s0 statements will have comments saying on which host the statement would be executed if you used \*(L"\-\-execute\*(R". .PP Technical side note: the first \s-1DSN\s0 is always the \*(L"left\*(R" server and the other DSNs are always the \*(L"right\*(R" server. Since either server can become the source or destination it's confusing to think of them as \*(L"src\*(R" and \*(L"dst\*(R". Therefore, they're generically referred to as left and right. It's easy to remember this because the first \s-1DSN\s0 is always to the left of the other server DSNs on the command line. .SH "EXIT STATUS" .IX Header "EXIT STATUS" The following are the exit statuses (also called return values, or return codes) when pt-table-sync finishes and exits. .PP .Vb 6 \& STATUS MEANING \& ====== ======================================================= \& 0 Success. \& 1 Internal error. \& 2 At least one table differed on the destination. \& 3 Combination of 1 and 2. .Ve .SH "OPTIONS" .IX Header "OPTIONS" Specify at least one of \*(L"\-\-print\*(R", \*(L"\-\-execute\*(R", or \*(L"\-\-dry\-run\*(R". .PP \&\*(L"\-\-where\*(R" and \*(L"\-\-replicate\*(R" are mutually exclusive. .PP This tool accepts additional command-line arguments. Refer to the \&\*(L"\s-1SYNOPSIS\*(R"\s0 and usage information for details. .IP "\-\-algorithms" 4 .IX Item "--algorithms" type: string; default: Chunk,Nibble,GroupBy,Stream .Sp Algorithm to use when comparing the tables, in order of preference. .Sp For each table, pt-table-sync will check if the table can be synced with the given algorithms in the order that they're given. The first algorithm that can sync the table is used. See \*(L"\s-1ALGORITHMS\*(R"\s0. .IP "\-\-ask\-pass" 4 .IX Item "--ask-pass" Prompt for a password when connecting to MySQL. .IP "\-\-bidirectional" 4 .IX Item "--bidirectional" Enable bidirectional sync between first and subsequent hosts. .Sp See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-[no]bin\-log" 4 .IX Item "--[no]bin-log" default: yes .Sp Log to the binary log (\f(CW\*(C`SET SQL_LOG_BIN=1\*(C'\fR). .Sp Specifying \f(CW\*(C`\-\-no\-bin\-log\*(C'\fR will \f(CW\*(C`SET SQL_LOG_BIN=0\*(C'\fR. .IP "\-\-buffer\-in\-mysql" 4 .IX Item "--buffer-in-mysql" Instruct MySQL to buffer queries in its memory. .Sp This option adds the \f(CW\*(C`SQL_BUFFER_RESULT\*(C'\fR option to the comparison queries. This causes MySQL to execute the queries and place them in a temporary table internally before sending the results back to pt-table-sync. The advantage of this strategy is that pt-table-sync can fetch rows as desired without using a lot of memory inside the Perl process, while releasing locks on the MySQL table (to reduce contention with other queries). The disadvantage is that it uses more memory on the MySQL server instead. .Sp You probably want to leave \*(L"\-\-[no]buffer\-to\-client\*(R" enabled too, because buffering into a temp table and then fetching it all into Perl's memory is probably a silly thing to do. This option is most useful for the GroupBy and Stream algorithms, which may fetch a lot of data from the server. .IP "\-\-[no]buffer\-to\-client" 4 .IX Item "--[no]buffer-to-client" default: yes .Sp Fetch rows one-by-one from MySQL while comparing. .Sp This option enables \f(CW\*(C`mysql_use_result\*(C'\fR which causes MySQL to hold the selected rows on the server until the tool fetches them. This allows the tool to use less memory but may keep the rows locked on the server longer. .Sp If this option is disabled by specifying \f(CW\*(C`\-\-no\-buffer\-to\-client\*(C'\fR then \&\f(CW\*(C`mysql_store_result\*(C'\fR is used which causes MySQL to send all selected rows to the tool at once. This may result in the results \*(L"cursor\*(R" being held open for a shorter time on the server, but if the tables are large, it could take a long time anyway, and use all your memory. .Sp For most non-trivial data sizes, you want to leave this option enabled. .Sp This option is disabled when \*(L"\-\-bidirectional\*(R" is used. .IP "\-\-channel" 4 .IX Item "--channel" type: string .Sp Channel name used when connected to a server using replication channels. Suppose you have two masters, master_a at port 12345, master_b at port 1236 and a slave connected to both masters using channels chan_master_a and chan_master_b. If you want to run pt-table-sync to syncronize the slave against master_a, pt-table-sync won't be able to determine what's the correct master since \s-1SHOW SLAVE STATUS\s0 will return 2 rows. In this case, you can use \-\-channel=chan_master_a to specify the channel name to use in the \s-1SHOW SLAVE STATUS\s0 command. .IP "\-\-charset" 4 .IX Item "--charset" short form: \-A; type: string .Sp Default character set. If the value is utf8, sets Perl's binmode on \&\s-1STDOUT\s0 to utf8, passes the mysql_enable_utf8 option to DBD::mysql, and runs \s-1SET NAMES UTF8\s0 after connecting to MySQL. Any other value sets binmode on \s-1STDOUT\s0 without the utf8 layer, and runs \s-1SET NAMES\s0 after connecting to MySQL. .IP "\-\-[no]check\-child\-tables" 4 .IX Item "--[no]check-child-tables" default: yes .Sp Check if \*(L"\-\-execute\*(R" will adversely affect child tables. When \&\*(L"\-\-replace\*(R", \*(L"\-\-replicate\*(R", or \*(L"\-\-sync\-to\-master\*(R" is specified, the tool may sync tables using \f(CW\*(C`REPLACE\*(C'\fR statements. If a table being synced has child tables with \f(CW\*(C`ON DELETE CASCADE\*(C'\fR, \f(CW\*(C`ON UPDATE CASCADE\*(C'\fR, or \f(CW\*(C`ON UPDATE SET NULL\*(C'\fR, the tool prints an error and skips the table because \&\f(CW\*(C`REPLACE\*(C'\fR becomes \f(CW\*(C`DELETE\*(C'\fR then \f(CW\*(C`INSERT\*(C'\fR, so the \f(CW\*(C`DELETE\*(C'\fR will cascade to the child table and delete its rows. In the worst case, this can delete all rows in child tables! .Sp Specify \f(CW\*(C`\-\-no\-check\-child\-tables\*(C'\fR to disable this check. To completely avoid affecting child tables, also specify \f(CW\*(C`\-\-no\-foreign\-key\-checks\*(C'\fR so MySQL will not cascade any operations from the parent to child tables. .Sp This check is only preformed if \*(L"\-\-execute\*(R" and one of \*(L"\-\-replace\*(R", \&\*(L"\-\-replicate\*(R", or \*(L"\-\-sync\-to\-master\*(R" is specified. \*(L"\-\-print\*(R" does not check child tables. .Sp The error message only prints the first child table found with an \&\f(CW\*(C`ON DELETE CASCADE\*(C'\fR, \f(CW\*(C`ON UPDATE CASCADE\*(C'\fR, or \f(CW\*(C`ON UPDATE SET NULL\*(C'\fR foreign key constraint. There could be other affected child tables. .IP "\-\-[no]check\-master" 4 .IX Item "--[no]check-master" default: yes .Sp With \*(L"\-\-sync\-to\-master\*(R", try to verify that the detected master is the real master. .IP "\-\-[no]check\-slave" 4 .IX Item "--[no]check-slave" default: yes .Sp Check whether the destination server is a slave. .Sp If the destination server is a slave, it's generally unsafe to make changes on it. However, sometimes you have to; \*(L"\-\-replace\*(R" won't work unless there's a unique index, for example, so you can't make changes on the master in that scenario. By default pt-table-sync will complain if you try to change data on a slave. Specify \f(CW\*(C`\-\-no\-check\-slave\*(C'\fR to disable this check. Use it at your own risk. .IP "\-\-[no]check\-triggers" 4 .IX Item "--[no]check-triggers" default: yes .Sp Check that no triggers are defined on the destination table. .Sp Triggers were introduced in MySQL v5.0.2, so for older versions this option has no effect because triggers will not be checked. .IP "\-\-chunk\-column" 4 .IX Item "--chunk-column" type: string .Sp Chunk the table on this column. .IP "\-\-chunk\-index" 4 .IX Item "--chunk-index" type: string .Sp Chunk the table using this index. .IP "\-\-chunk\-size" 4 .IX Item "--chunk-size" type: string; default: 1000 .Sp Number of rows or data size per chunk. .Sp The size of each chunk of rows for the \*(L"Chunk\*(R" and \*(L"Nibble\*(R" algorithms. The size can be either a number of rows, or a data size. Data sizes are specified with a suffix of k=kibibytes, M=mebibytes, G=gibibytes. Data sizes are converted to a number of rows by dividing by the average row length. .IP "\-\-columns" 4 .IX Item "--columns" short form: \-c; type: array .Sp Compare this comma-separated list of columns. .IP "\-\-config" 4 .IX Item "--config" type: Array .Sp Read this comma-separated list of config files; if specified, this must be the first option on the command line. .IP "\-\-conflict\-column" 4 .IX Item "--conflict-column" type: string .Sp Compare this column when rows conflict during a \*(L"\-\-bidirectional\*(R" sync. .Sp When a same but differing row is found the value of this column from each row is compared according to \*(L"\-\-conflict\-comparison\*(R", \*(L"\-\-conflict\-value\*(R" and \*(L"\-\-conflict\-threshold\*(R" to determine which row has the correct data and becomes the source. The column can be any type for which there is an appropriate \*(L"\-\-conflict\-comparison\*(R" (this is almost all types except, for example, blobs). .Sp This option only works with \*(L"\-\-bidirectional\*(R". See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-conflict\-comparison" 4 .IX Item "--conflict-comparison" type: string .Sp Choose the \*(L"\-\-conflict\-column\*(R" with this property as the source. .Sp The option affects how the \*(L"\-\-conflict\-column\*(R" values from the conflicting rows are compared. Possible comparisons are one of these MAGIC_comparisons: .Sp .Vb 1 \& newest|oldest|greatest|least|equals|matches \& \& COMPARISON CHOOSES ROW WITH \& ========== ========================================================= \& newest Newest temporal \-\-conflict\-column value \& oldest Oldest temporal \-\-conflict\-column value \& greatest Greatest numerical "\-\-conflict\-column value \& least Least numerical \-\-conflict\-column value \& equals \-\-conflict\-column value equal to \-\-conflict\-value \& matches \-\-conflict\-column value matching Perl regex pattern \& \-\-conflict\-value .Ve .Sp This option only works with \*(L"\-\-bidirectional\*(R". See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-conflict\-error" 4 .IX Item "--conflict-error" type: string; default: warn .Sp How to report unresolvable conflicts and conflict errors .Sp This option changes how the user is notified when a conflict cannot be resolved or causes some kind of error. Possible values are: .Sp .Vb 2 \& * warn: Print a warning to STDERR about the unresolvable conflict \& * die: Die, stop syncing, and print a warning to STDERR .Ve .Sp This option only works with \*(L"\-\-bidirectional\*(R". See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-conflict\-threshold" 4 .IX Item "--conflict-threshold" type: string .Sp Amount by which one \*(L"\-\-conflict\-column\*(R" must exceed the other. .Sp The \*(L"\-\-conflict\-threshold\*(R" prevents a conflict from being resolved if the absolute difference between the two \*(L"\-\-conflict\-column\*(R" values is less than this amount. For example, if two \*(L"\-\-conflict\-column\*(R" have timestamp values \*(L"2009\-12\-01 12:00:00\*(R" and \*(L"2009\-12\-01 12:05:00\*(R" the difference is 5 minutes. If \*(L"\-\-conflict\-threshold\*(R" is set to \*(L"5m\*(R" the conflict will be resolved, but if \*(L"\-\-conflict\-threshold\*(R" is set to \*(L"6m\*(R" the conflict will fail to resolve because the difference is not greater than or equal to 6 minutes. In this latter case, \*(L"\-\-conflict\-error\*(R" will report the failure. .Sp This option only works with \*(L"\-\-bidirectional\*(R". See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-conflict\-value" 4 .IX Item "--conflict-value" type: string .Sp Use this value for certain \*(L"\-\-conflict\-comparison\*(R". .Sp This option gives the value for \f(CW\*(C`equals\*(C'\fR and \f(CW\*(C`matches\*(C'\fR \&\*(L"\-\-conflict\-comparison\*(R". .Sp This option only works with \*(L"\-\-bidirectional\*(R". See \*(L"\s-1BIDIRECTIONAL SYNCING\*(R"\s0 for more information. .IP "\-\-databases" 4 .IX Item "--databases" short form: \-d; type: hash .Sp Sync only this comma-separated list of databases. .Sp A common request is to sync tables from one database with tables from another database on the same or different server. This is not yet possible. \&\*(L"\-\-databases\*(R" will not do it, and you can't do it with the D part of the \s-1DSN\s0 either because in the absence of a table name it assumes the whole server should be synced and the D part controls only the connection's default database. .IP "\-\-defaults\-file" 4 .IX Item "--defaults-file" short form: \-F; type: string .Sp Only read mysql options from the given file. You must give an absolute pathname. .IP "\-\-dry\-run" 4 .IX Item "--dry-run" Analyze, decide the sync algorithm to use, print and exit. .Sp Implies \*(L"\-\-verbose\*(R" so you can see the results. The results are in the same output format that you'll see from actually running the tool, but there will be zeros for rows affected. This is because the tool actually executes, but stops before it compares any data and just returns zeros. The zeros do not mean there are no changes to be made. .IP "\-\-engines" 4 .IX Item "--engines" short form: \-e; type: hash .Sp Sync only this comma-separated list of storage engines. .IP "\-\-execute" 4 .IX Item "--execute" Execute queries to make the tables have identical data. .Sp This option makes pt-table-sync actually sync table data by executing all the queries that it created to resolve table differences. Therefore, \fBthe tables will be changed!\fR And unless you also specify \*(L"\-\-verbose\*(R", the changes will be made silently. If this is not what you want, see \&\*(L"\-\-print\*(R" or \*(L"\-\-dry\-run\*(R". .IP "\-\-explain\-hosts" 4 .IX Item "--explain-hosts" Print connection information and exit. .Sp Print out a list of hosts to which pt-table-sync will connect, with all the various connection options, and exit. .IP "\-\-float\-precision" 4 .IX Item "--float-precision" type: int .Sp Precision for \f(CW\*(C`FLOAT\*(C'\fR and \f(CW\*(C`DOUBLE\*(C'\fR number-to-string conversion. Causes \s-1FLOAT\s0 and \s-1DOUBLE\s0 values to be rounded to the specified number of digits after the decimal point, with the \s-1\fBROUND\s0()\fR function in MySQL. This can help avoid checksum mismatches due to different floating-point representations of the same values on different MySQL versions and hardware. The default is no rounding; the values are converted to strings by the \s-1\fBCONCAT\s0()\fR function, and MySQL chooses the string representation. If you specify a value of 2, for example, then the values 1.008 and 1.009 will be rounded to 1.01, and will checksum as equal. .IP "\-\-[no]foreign\-key\-checks" 4 .IX Item "--[no]foreign-key-checks" default: yes .Sp Enable foreign key checks (\f(CW\*(C`SET FOREIGN_KEY_CHECKS=1\*(C'\fR). .Sp Specifying \f(CW\*(C`\-\-no\-foreign\-key\-checks\*(C'\fR will \f(CW\*(C`SET FOREIGN_KEY_CHECKS=0\*(C'\fR. .IP "\-\-function" 4 .IX Item "--function" type: string .Sp Which hash function you'd like to use for checksums. .Sp The default is \f(CW\*(C`CRC32\*(C'\fR. Other good choices include \f(CW\*(C`MD5\*(C'\fR and \f(CW\*(C`SHA1\*(C'\fR. If you have installed the \f(CW\*(C`FNV_64\*(C'\fR user-defined function, \f(CW\*(C`pt\-table\-sync\*(C'\fR will detect it and prefer to use it, because it is much faster than the built-ins. You can also use \s-1MURMUR_HASH\s0 if you've installed that user-defined function. Both of these are distributed with Percona Server. See pt-table-checksum for more information and benchmarks. .IP "\-\-help" 4 .IX Item "--help" Show help and exit. .IP "\-\-[no]hex\-blob" 4 .IX Item "--[no]hex-blob" default: yes .Sp \&\f(CW\*(C`HEX()\*(C'\fR \f(CW\*(C`BLOB\*(C'\fR, \f(CW\*(C`TEXT\*(C'\fR and \f(CW\*(C`BINARY\*(C'\fR columns. .Sp When row data from the source is fetched to create queries to sync the data (i.e. the queries seen with \*(L"\-\-print\*(R" and executed by \*(L"\-\-execute\*(R"), binary columns are wrapped in \s-1\fBHEX\s0()\fR so the binary data does not produce an invalid \s-1SQL\s0 statement. You can disable this option but you probably shouldn't. .IP "\-\-host" 4 .IX Item "--host" short form: \-h; type: string .Sp Connect to host. .IP "\-\-ignore\-columns" 4 .IX Item "--ignore-columns" type: Hash .Sp Ignore this comma-separated list of column names in comparisons. .Sp This option causes columns not to be compared. However, if a row is determined to differ between tables, all columns in that row will be synced, regardless. (It is not currently possible to exclude columns from the sync process itself, only from the comparison.) .IP "\-\-ignore\-databases" 4 .IX Item "--ignore-databases" type: Hash .Sp Ignore this comma-separated list of databases. .Sp (system databases such as \fBinformation_schema\fR and \fBperformance_schema\fR are ignored by default) .IP "\-\-ignore\-engines" 4 .IX Item "--ignore-engines" type: Hash; default: \s-1FEDERATED\s0,MRG_MyISAM .Sp Ignore this comma-separated list of storage engines. .IP "\-\-ignore\-tables" 4 .IX Item "--ignore-tables" type: Hash .Sp Ignore this comma-separated list of tables. .Sp Table names may be qualified with the database name. .IP "\-\-ignore\-tables\-regex" 4 .IX Item "--ignore-tables-regex" type: string; group: Filter .Sp Ignore tables whose names match the Perl regex. .IP "\-\-[no]index\-hint" 4 .IX Item "--[no]index-hint" default: yes .Sp Add \s-1FORCE/USE INDEX\s0 hints to the chunk and row queries. .Sp By default \f(CW\*(C`pt\-table\-sync\*(C'\fR adds a \s-1FORCE/USE INDEX\s0 hint to each \s-1SQL\s0 statement to coerce MySQL into using the index chosen by the sync algorithm or specified by \*(L"\-\-chunk\-index\*(R". This is usually a good thing, but in rare cases the index may not be the best for the query so you can suppress the index hint by specifying \f(CW\*(C`\-\-no\-index\-hint\*(C'\fR and let MySQL choose the index. .Sp This does not affect the queries printed by \*(L"\-\-print\*(R"; it only affects the chunk and row queries that \f(CW\*(C`pt\-table\-sync\*(C'\fR uses to select and compare rows. .IP "\-\-lock" 4 .IX Item "--lock" type: int .Sp Lock tables: 0=none, 1=per sync cycle, 2=per table, or 3=globally. .Sp This uses \f(CW\*(C`LOCK TABLES\*(C'\fR. This can help prevent tables being changed while you're examining them. The possible values are as follows: .Sp .Vb 12 \& VALUE MEANING \& ===== ======================================================= \& 0 Never lock tables. \& 1 Lock and unlock one time per sync cycle (as implemented \& by the syncing algorithm). This is the most granular \& level of locking available. For example, the Chunk \& algorithm will lock each chunk of C rows, and then \& unlock them if they are the same on the source and the \& destination, before moving on to the next chunk. \& 2 Lock and unlock before and after each table. \& 3 Lock and unlock once for every server (DSN) synced, with \& C. .Ve .Sp A replication slave is never locked if \*(L"\-\-replicate\*(R" or \*(L"\-\-sync\-to\-master\*(R" is specified, since in theory locking the table on the master should prevent any changes from taking place. (You are not changing data on your slave, right?) If \*(L"\-\-wait\*(R" is given, the master (source) is locked and then the tool waits for the slave to catch up to the master before continuing. .Sp If \f(CW\*(C`\-\-transaction\*(C'\fR is specified, \f(CW\*(C`LOCK TABLES\*(C'\fR is not used. Instead, lock and unlock are implemented by beginning and committing transactions. The exception is if \*(L"\-\-lock\*(R" is 3. .Sp If \f(CW\*(C`\-\-no\-transaction\*(C'\fR is specified, then \f(CW\*(C`LOCK TABLES\*(C'\fR is used for any value of \*(L"\-\-lock\*(R". See \*(L"\-\-[no]transaction\*(R". .IP "\-\-lock\-and\-rename" 4 .IX Item "--lock-and-rename" Lock the source and destination table, sync, then swap names. This is useful as a less-blocking \s-1ALTER TABLE,\s0 once the tables are reasonably in sync with each other (which you may choose to accomplish via any number of means, including dump and reload or even something like pt-archiver). It requires exactly two DSNs and assumes they are on the same server, so it does no waiting for replication or the like. Tables are locked with \s-1LOCK TABLES.\s0 .IP "\-\-password" 4 .IX Item "--password" short form: \-p; type: string .Sp Password to use when connecting. If password contains commas they must be escaped with a backslash: \*(L"exam\e,ple\*(R" .IP "\-\-pid" 4 .IX Item "--pid" type: string .Sp Create the given \s-1PID\s0 file. The tool won't start if the \s-1PID\s0 file already exists and the \s-1PID\s0 it contains is different than the current \s-1PID.\s0 However, if the \s-1PID\s0 file exists and the \s-1PID\s0 it contains is no longer running, the tool will overwrite the \s-1PID\s0 file with the current \s-1PID.\s0 The \s-1PID\s0 file is removed automatically when the tool exits. .IP "\-\-port" 4 .IX Item "--port" short form: \-P; type: int .Sp Port number to use for connection. .IP "\-\-print" 4 .IX Item "--print" Print queries that will resolve differences. .Sp If you don't trust \f(CW\*(C`pt\-table\-sync\*(C'\fR, or just want to see what it will do, this is a good way to be safe. These queries are valid \s-1SQL\s0 and you can run them yourself if you want to sync the tables manually. .IP "\-\-recursion\-method" 4 .IX Item "--recursion-method" type: array; default: processlist,hosts .Sp Preferred recursion method used to find slaves. .Sp Possible methods are: .Sp .Vb 5 \& METHOD USES \& =========== ================== \& processlist SHOW PROCESSLIST \& hosts SHOW SLAVE HOSTS \& none Do not find slaves .Ve .Sp The processlist method is preferred because \s-1SHOW SLAVE HOSTS\s0 is not reliable. However, the hosts method is required if the server uses a non-standard port (not 3306). Usually pt-table-sync does the right thing and finds the slaves, but you may give a preferred method and it will be used first. If it doesn't find any slaves, the other methods will be tried. .IP "\-\-replace" 4 .IX Item "--replace" Write all \f(CW\*(C`INSERT\*(C'\fR and \f(CW\*(C`UPDATE\*(C'\fR statements as \f(CW\*(C`REPLACE\*(C'\fR. .Sp This is automatically switched on as needed when there are unique index violations. .IP "\-\-replicate" 4 .IX Item "--replicate" type: string .Sp Sync tables listed as different in this table. .Sp Specifies that \f(CW\*(C`pt\-table\-sync\*(C'\fR should examine the specified table to find data that differs. The table is exactly the same as the argument of the same name to pt-table-checksum. That is, it contains records of which tables (and ranges of values) differ between the master and slave. .Sp For each table and range of values that shows differences between the master and slave, \f(CW\*(C`pt\-table\-checksum\*(C'\fR will sync that table, with the appropriate \f(CW\*(C`WHERE\*(C'\fR clause, to its master. .Sp This automatically sets \*(L"\-\-wait\*(R" to 60 and causes changes to be made on the master instead of the slave. .Sp If \*(L"\-\-sync\-to\-master\*(R" is specified, the tool will assume the server you specified is the slave, and connect to the master as usual to sync. .Sp Otherwise, it will try to use \f(CW\*(C`SHOW PROCESSLIST\*(C'\fR to find slaves of the server you specified. If it is unable to find any slaves via \f(CW\*(C`SHOW PROCESSLIST\*(C'\fR, it will inspect \f(CW\*(C`SHOW SLAVE HOSTS\*(C'\fR instead. You must configure each slave's \&\f(CW\*(C`report\-host\*(C'\fR, \f(CW\*(C`report\-port\*(C'\fR and other options for this to work right. After finding slaves, it will inspect the specified table on each slave to find data that needs to be synced, and sync it. .Sp The tool examines the master's copy of the table first, assuming that the master is potentially a slave as well. Any table that shows differences there will \&\fB\s-1NOT\s0\fR be synced on the slave(s). For example, suppose your replication is set up as A\->B, B\->C, B\->D. Suppose you use this argument and specify server B. The tool will examine server B's copy of the table. If it looks like server B's data in table \f(CW\*(C`test.tbl1\*(C'\fR is different from server A's copy, the tool will not sync that table on servers C and D. .IP "\-\-slave\-user" 4 .IX Item "--slave-user" type: string .Sp Sets the user to be used to connect to the slaves. This parameter allows you to have a different user with less privileges on the slaves but that user must exist on all slaves. .IP "\-\-slave\-password" 4 .IX Item "--slave-password" type: string .Sp Sets the password to be used to connect to the slaves. It can be used with \-\-slave\-user and the password for the user must be the same on all slaves. .IP "\-\-set\-vars" 4 .IX Item "--set-vars" type: Array .Sp Set the MySQL variables in this comma-separated list of \f(CW\*(C`variable=value\*(C'\fR pairs. .Sp By default, the tool sets: .Sp .Vb 1 \& wait_timeout=10000 .Ve .Sp Variables specified on the command line override these defaults. For example, specifying \f(CW\*(C`\-\-set\-vars wait_timeout=500\*(C'\fR overrides the defaultvalue of \f(CW10000\fR. .Sp The tool prints a warning and continues if a variable cannot be set. .IP "\-\-socket" 4 .IX Item "--socket" short form: \-S; type: string .Sp Socket file to use for connection. .IP "\-\-sync\-to\-master" 4 .IX Item "--sync-to-master" Treat the \s-1DSN\s0 as a slave and sync it to its master. .Sp Treat the server you specified as a slave. Inspect \f(CW\*(C`SHOW SLAVE STATUS\*(C'\fR, connect to the server's master, and treat the master as the source and the slave as the destination. Causes changes to be made on the master. Sets \*(L"\-\-wait\*(R" to 60 by default, sets \*(L"\-\-lock\*(R" to 1 by default, and disables \&\*(L"\-\-[no]transaction\*(R" by default. See also \*(L"\-\-replicate\*(R", which changes this option's behavior. .IP "\-\-tables" 4 .IX Item "--tables" short form: \-t; type: hash .Sp Sync only this comma-separated list of tables. .Sp Table names may be qualified with the database name. .IP "\-\-timeout\-ok" 4 .IX Item "--timeout-ok" Keep going if \*(L"\-\-wait\*(R" fails. .Sp If you specify \*(L"\-\-wait\*(R" and the slave doesn't catch up to the master's position before the wait times out, the default behavior is to abort. This option makes the tool keep going anyway. \fBWarning\fR: if you are trying to get a consistent comparison between the two servers, you probably don't want to keep going after a timeout. .IP "\-\-[no]transaction" 4 .IX Item "--[no]transaction" Use transactions instead of \f(CW\*(C`LOCK TABLES\*(C'\fR. .Sp The granularity of beginning and committing transactions is controlled by \&\*(L"\-\-lock\*(R". This is enabled by default, but since \*(L"\-\-lock\*(R" is disabled by default, it has no effect. .Sp Most options that enable locking also disable transactions by default, so if you want to use transactional locking (via \f(CW\*(C`LOCK IN SHARE MODE\*(C'\fR and \f(CW\*(C`FOR UPDATE\*(C'\fR, you must specify \f(CW\*(C`\-\-transaction\*(C'\fR explicitly. .Sp If you don't specify \f(CW\*(C`\-\-transaction\*(C'\fR explicitly \f(CW\*(C`pt\-table\-sync\*(C'\fR will decide on a per-table basis whether to use transactions or table locks. It currently uses transactions on InnoDB tables, and table locks on all others. .Sp If \f(CW\*(C`\-\-no\-transaction\*(C'\fR is specified, then \f(CW\*(C`pt\-table\-sync\*(C'\fR will not use transactions at all (not even for InnoDB tables) and locking is controlled by \*(L"\-\-lock\*(R". .Sp When enabled, either explicitly or implicitly, the transaction isolation level is set \f(CW\*(C`REPEATABLE READ\*(C'\fR and transactions are started \f(CW\*(C`WITH CONSISTENT SNAPSHOT\*(C'\fR. .IP "\-\-trim" 4 .IX Item "--trim" \&\f(CW\*(C`TRIM()\*(C'\fR \f(CW\*(C`VARCHAR\*(C'\fR columns in \f(CW\*(C`BIT_XOR\*(C'\fR and \f(CW\*(C`ACCUM\*(C'\fR modes. Helps when comparing MySQL 4.1 to >= 5.0. .Sp This is useful when you don't care about the trailing space differences between MySQL versions which vary in their handling of trailing spaces. MySQL 5.0 and later all retain trailing spaces in \f(CW\*(C`VARCHAR\*(C'\fR, while previous versions would remove them. .IP "\-\-[no]unique\-checks" 4 .IX Item "--[no]unique-checks" default: yes .Sp Enable unique key checks (\f(CW\*(C`SET UNIQUE_CHECKS=1\*(C'\fR). .Sp Specifying \f(CW\*(C`\-\-no\-unique\-checks\*(C'\fR will \f(CW\*(C`SET UNIQUE_CHECKS=0\*(C'\fR. .IP "\-\-user" 4 .IX Item "--user" short form: \-u; type: string .Sp User for login if not current user. .IP "\-\-verbose" 4 .IX Item "--verbose" short form: \-v; cumulative: yes .Sp Print results of sync operations. .Sp See \*(L"\s-1OUTPUT\*(R"\s0 for more details about the output. .IP "\-\-version" 4 .IX Item "--version" Show version and exit. .IP "\-\-[no]version\-check" 4 .IX Item "--[no]version-check" default: yes .Sp Check for the latest version of Percona Toolkit, MySQL, and other programs. .Sp This is a standard \*(L"check for updates automatically\*(R" feature, with two additional features. First, the tool checks its own version and also the versions of the following software: operating system, Percona Monitoring and Management (\s-1PMM\s0), MySQL, Perl, MySQL driver for Perl (DBD::mysql), and Percona Toolkit. Second, it checks for and warns about versions with known problems. For example, MySQL 5.5.25 had a critical bug and was re-released as 5.5.25a. .Sp A secure connection to Percona’s Version Check database server is done to perform these checks. Each request is logged by the server, including software version numbers and unique \s-1ID\s0 of the checked system. The \s-1ID\s0 is generated by the Percona Toolkit installation script or when the Version Check database call is done for the first time. .Sp Any updates or known problems are printed to \s-1STDOUT\s0 before the tool's normal output. This feature should never interfere with the normal operation of the tool. .Sp For more information, visit . .IP "\-\-wait" 4 .IX Item "--wait" short form: \-w; type: time .Sp How long to wait for slaves to catch up to their master. .Sp Make the master wait for the slave to catch up in replication before comparing the tables. The value is the number of seconds to wait before timing out (see also \*(L"\-\-timeout\-ok\*(R"). Sets \*(L"\-\-lock\*(R" to 1 and \*(L"\-\-[no]transaction\*(R" to 0 by default. If you see an error such as the following, .Sp .Vb 1 \& MASTER_POS_WAIT returned \-1 .Ve .Sp It means the timeout was exceeded and you need to increase it. .Sp The default value of this option is influenced by other options. To see what value is in effect, run with \*(L"\-\-help\*(R". .Sp To disable waiting entirely (except for locks), specify \*(L"\-\-wait\*(R" 0. This helps when the slave is lagging on tables that are not being synced. .IP "\-\-where" 4 .IX Item "--where" type: string .Sp \&\f(CW\*(C`WHERE\*(C'\fR clause to restrict syncing to part of the table. .IP "\-\-[no]zero\-chunk" 4 .IX Item "--[no]zero-chunk" default: yes .Sp Add a chunk for rows with zero or zero-equivalent values. The only has an effect when \*(L"\-\-chunk\-size\*(R" is specified. The purpose of the zero chunk is to capture a potentially large number of zero values that would imbalance the size of the first chunk. For example, if a lot of negative numbers were inserted into an unsigned integer column causing them to be stored as zeros, then these zero values are captured by the zero chunk instead of the first chunk and all its non-zero values. .SH "DSN OPTIONS" .IX Header "DSN OPTIONS" These \s-1DSN\s0 options are used to create a \s-1DSN.\s0 Each option is given like \&\f(CW\*(C`option=value\*(C'\fR. The options are case-sensitive, so P and p are not the same option. There cannot be whitespace before or after the \f(CW\*(C`=\*(C'\fR and if the value contains whitespace it must be quoted. \s-1DSN\s0 options are comma-separated. See the percona-toolkit manpage for full details. .IP "\(bu" 4 A .Sp dsn: charset; copy: yes .Sp Default character set. .IP "\(bu" 4 D .Sp dsn: database; copy: yes .Sp Database containing the table to be synced. .IP "\(bu" 4 F .Sp dsn: mysql_read_default_file; copy: yes .Sp Only read default options from the given file .IP "\(bu" 4 h .Sp dsn: host; copy: yes .Sp Connect to host. .IP "\(bu" 4 p .Sp dsn: password; copy: yes .Sp Password to use when connecting. If password contains commas they must be escaped with a backslash: \*(L"exam\e,ple\*(R" .IP "\(bu" 4 P .Sp dsn: port; copy: yes .Sp Port number to use for connection. .IP "\(bu" 4 S .Sp dsn: mysql_socket; copy: yes .Sp Socket file to use for connection. .IP "\(bu" 4 t .Sp copy: yes .Sp Table to be synced. .IP "\(bu" 4 u .Sp dsn: user; copy: yes .Sp User for login if not current user. .SH "ENVIRONMENT" .IX Header "ENVIRONMENT" The environment variable \f(CW\*(C`PTDEBUG\*(C'\fR enables verbose debugging output to \s-1STDERR.\s0 To enable debugging and capture all output to a file, run the tool like: .PP .Vb 1 \& PTDEBUG=1 pt\-table\-sync ... > FILE 2>&1 .Ve .PP Be careful: debugging output is voluminous and can generate several megabytes of output. .SH "ATTENTION" .IX Header "ATTENTION" Using <\s-1PTDEBUG\s0> might expose passwords. When debug is enabled, all command line parameters are shown in the output. .SH "SYSTEM REQUIREMENTS" .IX Header "SYSTEM REQUIREMENTS" You need Perl, \s-1DBI,\s0 DBD::mysql, and some core packages that ought to be installed in any reasonably new version of Perl. .SH "BUGS" .IX Header "BUGS" For a list of known bugs, see . .PP Please report bugs at . Include the following information in your bug report: .IP "\(bu" 4 Complete command-line used to run the tool .IP "\(bu" 4 Tool \*(L"\-\-version\*(R" .IP "\(bu" 4 MySQL version of all servers involved .IP "\(bu" 4 Output from the tool including \s-1STDERR\s0 .IP "\(bu" 4 Input files (log/dump/config files, etc.) .PP If possible, include debugging output by running the tool with \f(CW\*(C`PTDEBUG\*(C'\fR; see \*(L"\s-1ENVIRONMENT\*(R"\s0. .SH "DOWNLOADING" .IX Header "DOWNLOADING" Visit to download the latest release of Percona Toolkit. Or, get the latest release from the command line: .PP .Vb 1 \& wget percona.com/get/percona\-toolkit.tar.gz \& \& wget percona.com/get/percona\-toolkit.rpm \& \& wget percona.com/get/percona\-toolkit.deb .Ve .PP You can also get individual tools from the latest release: .PP .Vb 1 \& wget percona.com/get/TOOL .Ve .PP Replace \f(CW\*(C`TOOL\*(C'\fR with the name of any tool. .SH "AUTHORS" .IX Header "AUTHORS" Baron Schwartz .SH "ACKNOWLEDGMENTS" .IX Header "ACKNOWLEDGMENTS" My work is based in part on Giuseppe Maxia's work on distributed databases, and code derived from that article. There is more explanation, and a link to the code, at . .PP Another programmer extended Maxia's work even further. Fabien Coelho changed and generalized Maxia's technique, introducing symmetry and avoiding some problems that might have caused too-frequent checksum collisions. This work grew into pg_comparator, . Coelho also explained the technique further in a paper titled \*(L"Remote Comparison of Database Tables\*(R" (). .PP This existing literature mostly addressed how to find the differences between the tables, not how to resolve them once found. I needed a tool that would not only find them efficiently, but would then resolve them. I first began thinking about how to improve the technique further with my article , where I discussed a number of problems with the Maxia/Coelho \*(L"bottom-up\*(R" algorithm. After writing that article, I began to write this tool. I wanted to actually implement their algorithm with some improvements so I was sure I understood it completely. I discovered it is not what I thought it was, and is considerably more complex than it appeared to me at first. Fabien Coelho was kind enough to address some questions over email. .PP The first versions of this tool implemented a version of the Coelho/Maxia algorithm, which I called \*(L"bottom-up\*(R", and my own, which I called \*(L"top-down.\*(R" Those algorithms are considerably more complex than the current algorithms and I have removed them from this tool, and may add them back later. The improvements to the bottom-up algorithm are my original work, as is the top-down algorithm. The techniques to actually resolve the differences are also my own work. .PP Another tool that can synchronize tables is the SQLyog Job Agent from webyog. Thanks to Rohit Nadhani, \s-1SJA\s0's author, for the conversations about the general techniques. There is a comparison of pt-table-sync and \s-1SJA\s0 at .PP Thanks to the following people and organizations for helping in many ways: .PP The Rimm-Kaufman Group , MySQL \s-1AB\s0 , Blue Ridge InternetWorks , Percona , Fabien Coelho, Giuseppe Maxia and others at MySQL \s-1AB,\s0 Kristian Koehntopp (MySQL \s-1AB\s0), Rohit Nadhani (WebYog), The helpful monks at Perlmonks, And others too numerous to mention. .SH "ABOUT PERCONA TOOLKIT" .IX Header "ABOUT PERCONA TOOLKIT" This tool is part of Percona Toolkit, a collection of advanced command-line tools for MySQL developed by Percona. Percona Toolkit was forked from two projects in June, 2011: Maatkit and Aspersa. Those projects were created by Baron Schwartz and primarily developed by him and Daniel Nichter. Visit to learn about other free, open-source software from Percona. .SH "COPYRIGHT, LICENSE, AND WARRANTY" .IX Header "COPYRIGHT, LICENSE, AND WARRANTY" This program is copyright 2011\-2021 Percona \s-1LLC\s0 and/or its affiliates, 2007\-2011 Baron Schwartz. .PP \&\s-1THIS PROGRAM IS PROVIDED \*(L"AS IS\*(R" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.\s0 .PP This program is free software; you can redistribute it and/or modify it under the terms of the \s-1GNU\s0 General Public License as published by the Free Software Foundation, version 2; \s-1OR\s0 the Perl Artistic License. On \s-1UNIX\s0 and similar systems, you can issue `man perlgpl' or `man perlartistic' to read these licenses. .PP You should have received a copy of the \s-1GNU\s0 General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, \s-1MA\s0 02111\-1307 \s-1USA.\s0 .SH "VERSION" .IX Header "VERSION" pt-table-sync 3.4.0 .SH "POD ERRORS" .IX Header "POD ERRORS" Hey! \fBThe above document had some coding errors, which are explained below:\fR .IP "Around line 12836:" 4 .IX Item "Around line 12836:" Non-ASCII character seen before =encoding in 'Percona’s'. Assuming \s-1UTF\-8\s0