- NAME
- SYNOPSIS
- RISKS
- DESCRIPTION
- GROUP, MATCH AND KILL
- OUTPUT
- OPTIONS
- DSN OPTIONS
- DOWNLOADING
- ENVIRONMENT
- SYSTEM REQUIREMENTS
- BUGS
- COPYRIGHT, LICENSE AND WARRANTY
- AUTHOR
- ABOUT MAATKIT
- VERSION
mk-kill - Kill MySQL queries that match certain criteria.
Usage: mk-kill [OPTION]... [FILE...]
mk-kill kills MySQL connections. mk-kill connects to MySQL and gets queries from SHOW PROCESSLIST if no FILE is given. Else, it reads queries from one or more FILE which contains the output of SHOW PROCESSLIST. If FILE is -, mk-kill reads from STDIN.
Kill queries running longer than 60s:
mk-kill --busy-time 60 --kill
Print, do not kill, queries running longer than 60s:
mk-kill --busy-time 60 --print
Check for sleeping processes and kill them all every 10s:
mk-kill --match-command Sleep --kill --victims all --interval 10
Print all login processes:
mk-kill --match-state login --print --victims all
See which queries in the processlist right now would match:
mysql -e "SHOW PROCESSLIST" | mk-kill --busy-time 60 --print
The following section is included to inform users about the potential risks, whether known or unknown, of using this tool. The two main categories of risks are those created by the nature of the tool (e.g. read-only tools vs. read-write tools) and those created by bugs.
mk-kill is designed to kill queries if you use the --kill option is given, and that might disrupt your database's users, of course. You should test with the <--print> option, which is safe, if you're unsure what the tool will do.
At the time of this release, we know of no bugs that could cause serious harm to users.
The authoritative source for updated information is always the online issue tracking system. Issues that affect this tool will be marked as such. You can see a list of such issues at the following URL: http://www.maatkit.org/bugs/mk-kill.
See also BUGS for more information on filing bugs and getting help.
mk-kill captures queries from SHOW PROCESSLIST, filters them, and then either kills or prints them. This is also known as a "slow query sniper" in some circles. The idea is to watch for queries that might be consuming too many resources, and kill them.
For brevity, we talk about killing queries, but they may just be printed (or some other future action) depending on what options are given.
Normally mk-kill connects to MySQL to get queries from SHOW PROCESSLIST. Alternatively, it can read SHOW PROCESSLIST output from files. In this case, mk-kill does not connect to MySQL and --kill has no effect. You should use --print instead when reading files. The ability to read a file (or - for STDIN) allows you to capture SHOW PROCESSLIST and test it later with mk-kill to make sure that your matches kill the proper queries. There are a lot of special rules to follow, such as "don't kill replication threads," so be careful to not kill something important!
Two important options to know are --busy-time and --victims. First, whereas most match/filter options match their corresponding value from SHOW PROCESSLIST (e.g. --match-command matches a query's Command value), the Time value is matched by --busy-time. See also --interval.
Second, --victims controls which matching queries from each class are killed. By default, the matching query with the highest Time value is killed (the oldest query). See the next section, GROUP, MATCH AND KILL, for more details.
Usually you need to specify at least one --match option, else no
queries will match. Or, you can specify --match-all to match all queries
that aren't ignored by an --ignore option.
mk-kill is a work in progress, and there is much more it could do.
Queries pass through several steps to determine which exactly will be killed (or printed--whatever action is specified). Understanding these steps will help you match precisely the queries you want.
The first step is grouping queries into classes. The --group-by option controls grouping. By default, this option has no value so all queries are grouped into one, big default class. All types of matching and filtering (the next step) are applied per-class. Therefore, you may need to group queries in order to match/filter some classes but not others.
The second step is matching. Matching implies filtering since if a query
doesn't match some criteria, it is removed from its class.
Matching happens for each class. First, queries are filtered from their
class by the various Query Matches options like --match-user.
Then, entire classes are filtered by the various Class Matches options
like --query-count.
The third step is victim selection, that is, which matching queries in each class to kill. This is controlled by the --victims option. Although many queries in a class may match, you may only want to kill the oldest query, or all queries, etc.
The forth and final step is to take some action on all matching queries
from all classes. The Actions options specify which actions will be
taken. At this step, there are no more classes, just a single list of
queries to kill, print, etc.
If only --kill then there is no output. If only --print then a timestamped KILL statement if printed for every query that would have been killed, like:
# 2009-07-15T15:04:01 KILL 8 (Query 42 sec) SELECT * FROM huge_table
The line shows a timestamp, the query's Id (8), its Time (42 sec) and its Info (usually the query SQL).
If both --kill and --print are given, then matching queries are killed and a line for each like the one above is printed.
Any command executed by --execute-command is responsible for its own output and logging. After being executed, mk-kill has no control or interaction with the command.
Specify at least one of --kill, --kill-query, --print, --execute-command or --stop.
--any-busy-time and --each-busy-time are mutually exclusive.
--kill and --kill-query are mutually exclusive.
This tool accepts additional command-line arguments. Refer to the SYNOPSIS and usage information for details.
Prompt for a password when connecting to MySQL.
short form: -A; type: string
Default character set. If the value is utf8, sets Perl's binmode on STDOUT to utf8, passes the mysql_enable_utf8 option to DBD::mysql, and runs SET NAMES UTF8 after connecting to MySQL. Any other value sets binmode on STDOUT without the utf8 layer, and runs SET NAMES after connecting to MySQL.
type: Array
Read this comma-separated list of config files; if specified, this must be the first option on the command line.
Fork to the background and detach from the shell. POSIX operating systems only.
short form: -F; type: string
Only read mysql options from the given file. You must give an absolute pathname.
type: string
Apply matches to each class of queries grouped by this SHOW PROCESSLIST column.
In addition to the basic columns of SHOW PROCESSLIST (user, host, command,
state, etc.), queries can be matched by fingerprint which abstracts the
SQL query in the Info column.
By default, queries are not grouped, so matches and actions apply to all queries. Grouping allows matches and actions to apply to classes of similar queries, if any queries in the class match.
For example, detecting cache stampedes (see all-but-oldest under
--victims for an explanation of that term) requires that queries are
grouped by the arg attribute. This creates classes of identical queries
(stripped of comments). So queries "SELECT c FROM t WHERE id=1" and
"SELECT c FROM t WHERE id=1" are grouped into the same class, but
query c<"SELECT c FROM t WHERE id=3"> is not identical to the first two
queries so it is grouped into another class. Then when --victims
all-but-oldest is specified, all but the oldest query in each class is
killed for each class of queries that matches the match criteria.
Show help and exit.
short form: -h; type: string; default: localhost
Connect to host.
type: time
How often to check for queries to kill. If --busy-time is not given, then the default interval is 30 seconds. Else the default is half as often as --busy-time. If both --interval and --busy-time are given, then the explicit --interval value is used.
See also --run-time.
type: string
Print all output to this file when daemonized.
short form: -p; type: string
Password to use when connecting.
type: string
Create the given PID file when daemonized. The file contains the process ID of the daemonized instance. The PID file is removed when the daemonized instance exits. The program checks for the existence of the PID file when starting; if it exists and the process with the matching PID exists, the program exits.
short form: -P; type: int
Port number to use for connection.
default: yes
Remove SQL comments from queries in the Info column of the PROCESSLIST.
type: time
How long to run before exiting. By default mk-kill runs forever, or until its process is killed or stopped by the creation of a --sentinel file. If this option is specified, mk-kill runs for the specified amount of time and sleeps --interval seconds between each check of the PROCESSLIST.
type: string; default: /tmp/mk-kill-sentinel
Exit if this file exists.
The presence of the file specified by --sentinel will cause all running instances of mk-kill to exit. You might find this handy to stop cron jobs gracefully if necessary. See also --stop.
type: string; default: wait_timeout=10000
Set these MySQL variables. Immediately after connecting to MySQL, this string will be appended to SET and executed.
short form: -S; type: string
Socket file to use for connection.
Stop running instances by creating the --sentinel file.
Causes mk-kill to create the sentinel file specified by --sentinel and exit. This should have the effect of stopping all running instances which are watching the same sentinel file.
short form: -u; type: string
User for login if not current user.
Show version and exit.
type: string; default: oldest
Which of the matching queries in each class will be killed. After classes have been matched/filtered, this option specifies which of the matching queries in each class will be killed (or printed, etc.). The following values are possible:
Only kill the single oldest query. This is to prevent killing queries that aren't really long-running, they're just long-waiting. This sorts matching queries by Time and kills the one with the highest Time value.
Kill all queries in the class.
Kill all but the oldest query. This is the inverse of the oldest value.
This value can be used to prevent "cache stampedes", the condition where several identical queries are executed and create a backlog while the first query attempts to finish. Since all queries are identical, all but the first query are killed so that it can complete and populate the cache.
type: time
Wait after killing a query, before looking for more to kill. The purpose of this is to give blocked queries a chance to execute, so we don't kill a query that's blocking a bunch of others, and then kill the others immediately afterwards.
type: time
Wait before killing a query. The purpose of this is to give --execute-command a chance to see the matching query and gather other MySQL or system information before it's killed.
These options filter queries from their classes. If a query does not
match, it is removed from its class. The --ignore options take precedence.
The matches for command, db, host, etc. correspond to the columns returned
by SHOW PROCESSLIST: Command, db, Host, etc. All pattern matches are
case-sensitive by default, but they can be made case-insensitive by specifying
a regex pattern like (?i-xsm:select).
See also GROUP, MATCH AND KILL.
group: Query Matches
Match all queries that are not ignored. If no ignore options are specified,
then every query matches (except replication threads, unless
--replication-threads is also specified). This option allows you to
specify negative matches, i.e. "match every query except..." where the
exceptions are defined by specifying various --ignore options.
This option is not the same as --victims all. This option matches
all queries within a class, whereas --victims all specifies that all
matching queries in a class (however they matched) will be killed. Normally,
however, the two are used together because if, for example, you specify
--victims oldest, then although all queries may match, only the oldest
will be killed.
type: time; group: Query Matches
Match queries that have been running for longer than this time. The queries must be in Command=Query status. This matches a query's Time value as reported by SHOW PROCESSLIST.
type: time; group: Query Matches
Match queries that have been idle/sleeping for longer than this time. The queries must be in Command=Sleep status. This matches a query's Time value as reported by SHOW PROCESSLIST.
type: string; group: Query Matches
Ignore queries whose Command matches this Perl regex.
See --match-command.
type: string; group: Query Matches
Ignore queries whose db (database) matches this Perl regex.
See --match-db.
type: string; group: Query Matches
Ignore queries whose Host matches this Perl regex.
See --match-host.
type: string; group: Query Matches
Ignore queries whose Info (query) matches this Perl regex.
See --match-info.
default: yes; group: Query Matches
Don't kill mk-kill's own connection.
type: string; group: Query Matches; default: Locked
Ignore queries whose State matches this Perl regex. The default is to keep threads from being killed if they are locked waiting for another thread.
See --match-state.
type: string; group: Query Matches
Ignore queries whose user matches this Perl regex.
See --match-user.
type: string; group: Query Matches
Match only queries whose Command matches this Perl regex.
Common Command values are:
Query Sleep Binlog Dump Connect Delayed insert Execute Fetch Init DB Kill Prepare Processlist Quit Reset stmt Table Dump
See http://dev.mysql.com/doc/refman/5.1/en/thread-commands.html for a full list and description of Command values.
type: string; group: Query Matches
Match only queries whose db (database) matches this Perl regex.
type: string; group: Query Matches
Match only queries whose Host matches this Perl regex.
The Host value often time includes the port like "host:port".
type: string; group: Query Matches
Match only queries whose Info (query) matches this Perl regex.
The Info column of the processlist shows the query that is being executed or NULL if no query is being executed.
type: string; group: Query Matches
Match only queries whose State matches this Perl regex.
Common State values are:
Locked login copy to tmp table Copying to tmp table Copying to tmp table on disk Creating tmp table executing Reading from net Sending data Sorting for order Sorting result Table lock Updating
See http://dev.mysql.com/doc/refman/5.1/en/general-thread-states.html for a full list and description of State values.
type: string; group: Query Matches
Match only queries whose User matches this Perl regex.
group: Query Matches
Allow matching and killing replication threads.
By default, matches do not apply to replication threads; i.e. replication threads are completely ignored. Specifying this option allows matches to match (and potentially kill) replication threads on masters and slaves.
These matches apply to entire query classes. Classes are created by specifying the --group-by option, else all queries are members of a single, default class.
See also GROUP, MATCH AND KILL.
type: time; group: Class Matches
Match query class if any query has been running for longer than this time.
"Longer than" means that if you specify 10, for example, the class will
only match if there's at least one query that has been running for greater
than 10 seconds.
See --each-busy-time for more details.
type: time; group: Class Matches
Match query class if each query has been running for longer than this time.
"Longer than" means that if you specify 10, for example, the class will
only match if each and every query has been running for greater than 10
seconds.
See also --any-busy-time (to match a class if ANY query has been running longer than the specified time) and --busy-time.
type: int; group: Class Matches
Match query class if it has at least this many queries. When queries are grouped into classes by specifying --group-by, this option causes matches to apply only to classes with at least this many queries. If --group-by is not specified then this option causes matches to apply only if there are at least this many queries in the entire SHOW PROCESSLIST.
short form: -v
Print information to STDOUT about what is being done.
These actions are taken for every matching query from all classes. The actions are taken in this order: --print, --execute-command, --kill/--kill-query. This order allows --execute-command to see the output of --print and the query before --kill/--kill-query. This may be helpful because mk-kill does not pass any information to --execute-command.
See also GROUP, MATCH AND KILL.
type: string; group: Actions
Execute this command when a query matches.
After the command is executed, mk-kill has no control over it, so the command is responsible for its own info gathering, logging, interval, etc. The command is executed each time a query matches, so be careful that the command behaves well when multiple instances are ran. No information from mk-kill is passed to the command.
See also --wait-before-kill.
group: Actions
Kill the connection for matching queries.
This option makes mk-kill kill the connections (a.k.a. processes, threads) that have matching queries. Use --kill-query if you only want to kill individual queries and not their connections.
Unless --print is also given, no other information is printed that shows that mk-kill matched and killed a query.
See also --wait-before-kill and --wait-after-kill.
group: Actions
Kill matching queries.
This option makes mk-kill kill matching queries. This requires MySQL 5.0 or newer. Unlike --kill which kills the connection for matching queries, this option only kills the query, not its connection.
group: Actions
Print a KILL statement for matching queries; does not actually kill queries.
If you just want to see which queries match and would be killed without actually killing them, specify --print. To both kill and print matching queries, specify both --kill and --print.
These DSN options are used to create a DSN. Each option is given like
option=value. The options are case-sensitive, so P and p are not the
same option. There cannot be whitespace before or after the = and
if the value contains whitespace it must be quoted. DSN options are
comma-separated. See the maatkit manpage for full details.
dsn: charset; copy: yes
Default character set.
dsn: database; copy: yes
Default database.
dsn: mysql_read_default_file; copy: yes
Only read default options from the given file
dsn: host; copy: yes
Connect to host.
dsn: password; copy: yes
Password to use when connecting.
dsn: port; copy: yes
Port number to use for connection.
dsn: mysql_socket; copy: yes
Socket file to use for connection.
dsn: user; copy: yes
User for login if not current user.
You can download Maatkit from Google Code at http://code.google.com/p/maatkit/, or you can get any of the tools easily with a command like the following:
wget http://www.maatkit.org/get/toolname or wget http://www.maatkit.org/trunk/toolname
Where toolname can be replaced with the name (or fragment of a name) of any
of the Maatkit tools. Once downloaded, they're ready to run; no installation is
needed. The first URL gets the latest released version of the tool, and the
second gets the latest trunk code from Subversion.
The environment variable MKDEBUG enables verbose debugging output in all of
the Maatkit tools:
MKDEBUG=1 mk-....
You need Perl, DBI, DBD::mysql, and some core packages that ought to be installed in any reasonably new version of Perl.
For a list of known bugs see http://www.maatkit.org/bugs/mk-kill.
Please use Google Code Issues and Groups to report bugs or request support: http://code.google.com/p/maatkit/. You can also join #maatkit on Freenode to discuss Maatkit.
Please include the complete command-line used to reproduce the problem you are
seeing, the version of all MySQL servers involved, the complete output of the
tool when run with --version, and if possible, debugging output produced by
running with the MKDEBUG=1 environment variable.
This program is copyright 2009-2011 Baron Schwartz. Feedback and improvements are welcome.
THIS PROGRAM IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 2; OR the Perl Artistic License. On UNIX and similar systems, you can issue `man perlgpl' or `man perlartistic' to read these licenses.
You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA.
Baron Schwartz, Daniel Nichter
This tool is part of Maatkit, a toolkit for power users of MySQL. Maatkit was created by Baron Schwartz; Baron and Daniel Nichter are the primary code contributors. Both are employed by Percona. Financial support for Maatkit development is primarily provided by Percona and its clients.
This manual page documents Ver 0.9.10 Distrib 7540 $Revision: 7531 $.