format of mysql query log

In the MySQL General Query Log, the fields are these:

yymmdd hh:mm:ss thread_id command_type query_body

The timestamp only appears each time it changes.

In your example "133" would be the thread id of the connected client, as shown in SHOW PROCESSLIST;.

Each line represents a query issued, but not necessarily a round-trip from the client to the database, because queries issued within stored procedures and stored functions are also logged, and a proc or function, while running, could issue many queries in response to a single query from a client.

Also, the MySQL C-API, which is the library underlying some languages' connectors to MySQL, supports multiple statement execution, which would be another case where multiple queries don't necessarily represent multiple round-trips since multiple queries can be send to the database as a "batch"... but that's only a factor if this capability is used by the calling application.

The general log can also be directed to write to the mysql.general_log table, which is easier to parse and analyze than the file-based log, but the same general observations, above, apply when using that format.

The slow query log includes more information and so provides a more informative log for understanding how the server is performing, and can be used as a log of all queries by configuring MySQL to interpret all queries as meeting the "slow" criteria. This is done by setting long_query_time to '0' after enabling the slow query log.

A typical general log format looks like this:

Time              Id Command    Argument
110208 12:12:06   771 Query rollback

Multiple queries can appear at particular time-stamp also ID indicates MySQL connection thread id which has executed a query Command of type "Query" which indicates query has been executed and Argument is actually a query itself.