MySQL Utilities

6.3.2 mysql.utilities.command.proc — Search Processes on Servers

This module searches processes on a server and optionally kills either the query or the connection for all matches.

Processes are matched by searching the fields of the INFORMATION_SCHEMA.PROCESSLIST table. Internally, the module operates by constructing a SELECT statement for finding matching processes, and then sending it to the server. Instead of performing the search, the module can return the SQL code that performs the query. This can be useful if you want to execute the query later or feed it to some other program that processes SQL queries further.


The following constants correspond to fields in the INFORMATION_SCHEMA.PROCESSLIST table. They indicate which columns to examine when searching for processes matching the search conditions.

The following constants indicate actions to perform on processes that match the search conditions.


class mysql.utilities.command.proc.ProcessGrep(matches, actions=[], use_regexp=False)

This class searches the INFORMATION_SCHEMA.PROCESSLIST table for processes on MySQL servers and optionally kills them. It can be used to perform a search (and optionally kill), or to generate the SQL statement for doing the same.

For example, to kill all queries with user 'mats', use the following code:

>>> from mysql.utilities.command.proc import *
>>> grep = ProcessGrep(matches=[(USER, "mats")], actions=[KILL_QUERY])
>>> grep.execute("", "")



Return the SQL code for executing the search (and optionally, the kill).

If only_body is True, only the body of the function is shown. This is useful if the SQL code is to be used with other utilities that generate the routine declaration. If only_body is False, a complete procedure is generated if there is any kill action supplied, and just a SELECT statement if it is a plain search.



SQL code for executing the operation specified by the options.

Return type:


execute(connections, ...[, output=sys.stdout, connector=mysql.connector])

Execute the search on each of the connections supplied. If output is not None, the value is treated as a file object and the result of the execution is printed on that stream. Note that the output and connector arguments must be supplied as keyword arguments. All other arguments are treated as connection specifiers.