Commit b4883a8044af6d35875dd8300a5d326f5460c729
1 parent
b7ded5be
Exists in
master
and in
1 other branch
installed query analyzer
Showing
3 changed files
with
151 additions
and
0 deletions
Show diff stats
| ... | ... | @@ -0,0 +1,96 @@ |
| 1 | += Query Analyzer Plugin for MySQL on Rails | |
| 2 | + | |
| 3 | +MODIFIED by John Eberly originally take from http://svn.nfectio.us | |
| 4 | + | |
| 5 | +The Query Analyzer plugin will expand the usability of your log files | |
| 6 | +by providing query analysis using the MySQL query execution plan. Each SQL | |
| 7 | +select query will be 'EXPLAIN'ed and added to the log files right below | |
| 8 | +the original query. | |
| 9 | + | |
| 10 | +Using this plugin and a good understanding of the results, you will be | |
| 11 | +able to analyze and optimize the queries your application is making. | |
| 12 | + | |
| 13 | +Refer to http://www.mysql.org/doc/refman/5.0/en/explain.html for more | |
| 14 | +information on understanding the results. | |
| 15 | + | |
| 16 | += Installation | |
| 17 | + | |
| 18 | +script/plugin install git://github.com/jeberly/query-analyzer.git | |
| 19 | + | |
| 20 | += Example Use | |
| 21 | + | |
| 22 | +Here is a real life usage of the plugin that detected the omission of indexes on | |
| 23 | +a table. In this case, it was a join table and the keys didn't have indexes (silly me!). | |
| 24 | +Names have been changed to protect the innocent (and make it fit 80 columns) | |
| 25 | + | |
| 26 | +# development.log | |
| 27 | + | |
| 28 | +P Load (0.008669) | |
| 29 | + => SELECT p.* FROM p INNER JOIN d ON p.id = d.p_id WHERE (d.p_id = 2 AND ((d.type = 'P'))) | |
| 30 | + | |
| 31 | +Analyzing P Load | |
| 32 | + | |
| 33 | +select_type | key_len | type | Extra | id | possible_keys | rows | table | ref | key | |
| 34 | +---------------------------------------------------------------------------------------------------- | |
| 35 | +SIMPLE | | ALL | Using where | 1 | | 74 | d | | | |
| 36 | +SIMPLE | 4 | eq_ref | Using where | 1 | PRIMARY | 1 | p | d.p_id | PRIMARY | |
| 37 | + | |
| 38 | + | |
| 39 | += Analyze the results | |
| 40 | + | |
| 41 | +Looking at the results of the execution plan, we can see that the lookup in | |
| 42 | +the d table is missing an index (possible_keys=null) and performed a full | |
| 43 | +table scan (type=ALL) to satisfy the WHERE condition. In this case, there was only one | |
| 44 | +row that matched the condition in the table, but MySQL still had to search all 74 rows | |
| 45 | +in the table to find it, a key indicator of a missing | |
| 46 | +or malformed index(es). Once it has pulled all the records to satisfy the WHERE, it then | |
| 47 | +starts the p table join. This time, it was able to match d.p_id to p.id using | |
| 48 | +the PRIMARY key on the p table. The type=eq_ref indicates a 1 to 1 match against a primary | |
| 49 | +or unique column. | |
| 50 | + | |
| 51 | +Lets add some indexes to the join table and see if we can cut that full table scan down in the | |
| 52 | +number of rows it needs to search. | |
| 53 | + | |
| 54 | +#> script/generate Migration AddIndexesToD | |
| 55 | + | |
| 56 | +# file: 005_add_indexes_to_d | |
| 57 | +class AddIndexesToD < ActiveRecord::Migration | |
| 58 | + def self.up | |
| 59 | + add_index :d, [ :p_id, :type ] | |
| 60 | + add_index :d, :type | |
| 61 | + end | |
| 62 | + | |
| 63 | + def self.remove | |
| 64 | + remove_index :d, [ :p_id, :type ] | |
| 65 | + remove_index :d, :type | |
| 66 | + end | |
| 67 | +end | |
| 68 | + | |
| 69 | + | |
| 70 | +Now that we have an index on the foreign_key column and type, lets re-run the query and | |
| 71 | +see if we got rid of that full table scan. | |
| 72 | + | |
| 73 | +# development.log | |
| 74 | + | |
| 75 | +P Load (0.009011) | |
| 76 | + => SELECT p.* FROM p INNER JOIN d ON p.id = d.p_id WHERE (d.p_id = 2 AND ((d.type = 'P'))) | |
| 77 | + | |
| 78 | +Analyzing P Load | |
| 79 | + | |
| 80 | +select_type | key_len | type | Extra | id | possible_keys | rows | table | ref | key | |
| 81 | +------------------------------------------------------------------------------------------------------------------------------ | |
| 82 | +SIMPLE | 255 | ref | Using where | 1 | d_p_id_type_index,d_type_index | 1 | d | const | d_p_id_type_index | |
| 83 | +SIMPLE | 4 | eq_ref | Using where | 1 | PRIMARY | 1 | p | d.p_id | PRIMARY | |
| 84 | + | |
| 85 | +Okay. Now MySQL is using an index satisfy the WHERE condition. Using the index, it was able to | |
| 86 | +find the single row that matched, preventing the full table scan. | |
| 87 | + | |
| 88 | + | |
| 89 | +Credits: | |
| 90 | +The extension of the Array class for printing the columnized records was originally | |
| 91 | +written by Peter Cooper who adapted it from Courtenay from #caboose. | |
| 92 | + | |
| 93 | +http://www.rubyinside.com/columnized-text-datasets-in-rails-71.html | |
| 94 | +http://habtm.com/articles/2006/06/10/pretty-tables-for-ruby-objects | |
| 95 | + | |
| 96 | +Released under the MIT license (download your own if you need it) | ... | ... |
| ... | ... | @@ -0,0 +1,54 @@ |
| 1 | +class Array | |
| 2 | + protected | |
| 3 | + def qa_columnized_row(fields, sized) | |
| 4 | + row = [] | |
| 5 | + fields.each_with_index do |f, i| | |
| 6 | + row << sprintf("%0-#{sized[i]}s", f.to_s) | |
| 7 | + end | |
| 8 | + row.join(' | ') | |
| 9 | + end | |
| 10 | + | |
| 11 | + public | |
| 12 | + | |
| 13 | + def qa_columnized | |
| 14 | + sized = {} | |
| 15 | + self.each do |row| | |
| 16 | + row.values.each_with_index do |value, i| | |
| 17 | + sized[i] = [sized[i].to_i, row.keys[i].length, value.to_s.length].max | |
| 18 | + end | |
| 19 | + end | |
| 20 | + | |
| 21 | + table = [] | |
| 22 | + table << qa_columnized_row(self.first.keys, sized) | |
| 23 | + table << '-' * table.first.length | |
| 24 | + self.each { |row| table << qa_columnized_row(row.values, sized) } | |
| 25 | + table.join("\n ") # Spaces added to work with format_log_entry | |
| 26 | + end | |
| 27 | +end | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | +module ActiveRecord | |
| 32 | + module ConnectionAdapters | |
| 33 | + class MysqlAdapter < AbstractAdapter | |
| 34 | + private | |
| 35 | + alias_method :select_without_analyzer, :select | |
| 36 | + | |
| 37 | + def select(sql, name = nil) | |
| 38 | + query_results = select_without_analyzer(sql, name) | |
| 39 | + | |
| 40 | + if @logger and @logger.level == Logger::DEBUG | |
| 41 | + @logger.debug( | |
| 42 | + ActiveRecord::Base.silence do | |
| 43 | + explain_results = select_without_analyzer("explain #{sql}", name) | |
| 44 | + format_log_entry("\033[1;34m############ FIXME - UNOPTIMIZED QUERY for #{name} ############ \033[0m\n", | |
| 45 | + "#{explain_results.qa_columnized}\n" | |
| 46 | + ) if explain_results[0]["rows"].to_i > 100 && sql =~ / where[\s(]/i | |
| 47 | + end | |
| 48 | + ) if sql =~ /^select/i | |
| 49 | + end | |
| 50 | + query_results | |
| 51 | + end | |
| 52 | + end | |
| 53 | + end | |
| 54 | +end | ... | ... |