The script filters the MySQL Slow Query Log to show queries which impacted performance most and is intended to be used by DB admins and application developers.

The log file is analyzed and processed as a stream, line after line, so there is no need to load the whole log file into memory.

--no-duplicates is a very useful option to see only necessary statistics.

--incremental remembers last input file positions and statistics in a SQLite 3 database, so periodical executions on the same files run much faster.

The Python version is usually 3-5 times faster than the PHP5 version.

You are welcome to contribute a version translated into your favorite scripting language, just open a feature request under Issues i.e. to add a Perl version.

# Filter slow queries executed for at least 3 seconds not from root, remove duplicates,
# apply execution count as first sorting value and save first 10 unique queries to file.
# In addition, remember last input file position and statistics.
php mysql_filter_slow_log.php -T=3 -eu=root --no-duplicates --sort-execution-count --top=10 --incremental linux-slow.log > mysql-slow-queries.log

# Start permanent filtering of all slow queries from now on: at least 3 seconds or examining 10000 rows, exclude users root and test
tail -f -n 0 linux-slow.log | python -T=3 -R=10000 -eu=root -eu=test &
# (-n 0 outputs only lines generated after start of tail)

# Stop permanent filtering
kill `ps auxww | grep 'tail -f -n 0 linux-slow.log' | egrep -v grep | awk '{print $2}'`


-ih, --include-host
-eh, --exclude-host
-iu, --include-user
-eu, --exclude-user
-iq, --include-query

--date=date_first-date_last Include only queries between date_first (and date_last).
                            Input:                    Date Range:
                            13.11.2006             -> 13.11.2006 - 14.11.2006 (exclusive)
                            13.11.2006-15.11.2006  -> 13.11.2006 - 16.11.2006 (exclusive)
                            15-11-2006-11/13/2006  -> 13.11.2006 - 16.11.2006 (exclusive)
                            >13.11.2006            -> 14.11.2006 - later
                            13.11.2006-            -> 13.11.2006 - later
                            <13.11.2006            -> earlier    - 13.11.2006 (exclusive)
                            -13.11.2006            -> earlier    - 14.11.2006 (exclusive)
                            Please do not forget to escape the greater or lesser than symbols (><, i.e. '--date=>13.11.2006').
                            Short dates are supported if you include a trailing separator (i.e. 13.11.-11/15/).

--incremental Remember input file positions and optionally --no-duplicates statistics between executions in mysql_filter_slow_log.sqlite3

--no-duplicates Powerful option to output only unique query strings with additional statistics:
                Execution count, first and last timestamp.
                Query time: avg / max / sum.
                Lock time: avg / max / sum.
                Rows examined: avg / max / sum.
                Rows sent: avg / max / sum.

--no-output Do not print statistics, just update database with incremental statistics

Default ordering of unique queries:
--sort-sum-query-time    [ 1. position]
--sort-avg-query-time    [ 2. position]
--sort-max-query-time    [ 3. position]
--sort-sum-lock-time     [ 4. position]
--sort-avg-lock-time     [ 5. position]
--sort-max-lock-time     [ 6. position]
--sort-sum-rows-examined [ 7. position]
--sort-avg-rows-examined [ 8. position]
--sort-max-rows-examined [ 9. position]
--sort-execution-count   [10. position]
--sort-sum-rows-sent     [11. position]
--sort-avg-rows-sent     [12. position]
--sort-max-rows-sent     [13. position]

--sort=sum-query-time,avg-query-time,max-query-time,...   You can include multiple sorting values separated by commas.
--sort=sqt,aqt,mqt,slt,alt,mlt,sre,are,mre,ec,srs,ars,mrs Every long sorting option has an equivalent short form (first character of each word).

--top=max_unique_query_count Output maximal max_unique_query_count different unique queries
--details                    Enables output of timestamp based unique query time lines after user list
                             (i.e. # Query_time: 81  Lock_time: 0  Rows_sent: 884  Rows_examined: 2448350).

--help Output this message only and quit

[multiple] options can be passed more than once to set multiple values.
[position] options take the position of their first occurrence into account.
           The first passed option will replace the default first sorting, ...
           Remaining default ordering options will keep their relative positions.

# I.e. you could add the following lines under the [mysqld] section of your my.ini or my.cnf configuration file:

# Log all queries taking more than 3 seconds
long_query_time=3  # minimum: 1, default: 10

# MySQL >= 5.1.21 (or patched): 3 seconds = 3000000 microseconds
# long_query_time=3.000000  # minimum: 0.000001 (1 microsecond)

# Activate the Slow Query Log
slow_query_log  # >= 5.1.29
# log-slow-queries  # deprecated since 5.1.29

# Write to a custom file name (>= 5.1.29)
# slow_query_log_file=file_name  # default: /data_dir/host_name-slow.log

# Log all queries without indexes
# log-queries-not-using-indexes

# Log only queries which examine at least N rows (>= 5.1.21)
# min_examined_row_limit=1000  # default: 0

# log-slow-admin-statements

# Log slow queries executed by replication slaves (>= 5.1.21)
# log-slow-slave-statements

# MySQL 5.1.6 through 5.1.20 had a default value of log-output=TABLE, so you should force
# Attention: logging to TABLE only includes whole seconds information

## Admin query for online activation is possible since MySQL 5.1 (without server restart)
## SET @@global.slow_query_log=1
## SET @@global.long_query_time=1

## Show current variables related to the Slow Query Log
## SHOW GLOBAL VARIABLES WHERE Variable_name REGEXP 'admin|min_examined|log_output|log_queries|log_slave|long|slow_quer'