Class: Dexter::Indexer

Inherits:
Object
  • Object
show all
Includes:
Logging
Defined in:
lib/dexter/indexer.rb

Constant Summary

Constants included from Logging

Logging::COLOR_CODES

Instance Method Summary collapse

Methods included from Logging

#colorize, #log, #output

Constructor Details

#initialize(options) ⇒ Indexer

Returns a new instance of Indexer.



5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
 
# File 'lib/dexter/indexer.rb', line 5

def initialize(options)
  @create = options[:create]
  @tablespace = options[:tablespace]
  @log_level = options[:log_level]
  @exclude_tables = options[:exclude]
  @include_tables = Array(options[:include].split(",")) if options[:include]
  @log_sql = options[:log_sql]
  @log_explain = options[:log_explain]
  @min_time = options[:min_time] || 0
  @min_calls = options[:min_calls] || 0
  @analyze = options[:analyze]
  @min_cost_savings_pct = options[:min_cost_savings_pct].to_i
  @log_table = options[:log_table]
  @options = options
  @mutex = Mutex.new

  create_extension unless extension_exists?
  execute("SET lock_timeout = '5s'")
end

Instance Method Details

#csvlog_activity(last_log_time) ⇒ Object

works with file_fdw: www.postgresql.org/docs/current/file-fdw.html log_fdw: docs.aws.amazon.com/AmazonRDS/latest/UserGuide/Appendix.PostgreSQL.CommonDBATasks.Extensions.foreign-data-wrappers.html



51
52
53
54
55
56
57
58
59
60
61
62
63
 
# File 'lib/dexter/indexer.rb', line 51

def csvlog_activity(last_log_time)
  query = <<~SQL
    SELECT
      log_time,
      message,
      detail
    FROM
      #{conn.quote_ident(@log_table)}
    WHERE
      log_time >= \$1
  SQL
  execute(query, params: [last_log_time])
end

#process_queries(queries) ⇒ Object 



78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
 
# File 'lib/dexter/indexer.rb', line 78

def process_queries(queries)
  # reset hypothetical indexes
  reset_hypothetical_indexes

  tables = Set.new(database_tables + materialized_views)

  # map tables without schema to schema
  no_schema_tables = {}
  search_path_index = Hash[search_path.map.with_index.to_a]
  tables.group_by { |t| t.split(".")[-1] }.each do |group, t2|
    no_schema_tables[group] = t2.sort_by { |t| [search_path_index[t.split(".")[0]] || 1000000, t] }[0]
  end

  # add tables from views
  view_tables = database_view_tables
  view_tables.each do |v, vt|
    view_tables[v] = vt.map { |t| no_schema_tables[t] || t }
  end

  # fully resolve tables
  # make sure no views in result
  view_tables.each do |v, vt|
    view_tables[v] = vt.flat_map { |t| view_tables[t] || [t] }.uniq
  end

  # filter queries from other databases and system tables
  queries.each do |query|
    # add schema to table if needed
    query.tables = query.tables.map { |t| no_schema_tables[t] || t }

    # substitute view tables
    new_tables = query.tables.flat_map { |t| view_tables[t] || [t] }.uniq
    query.tables_from_views = new_tables - query.tables
    query.tables = new_tables

    # check for missing tables
    query.missing_tables = !query.tables.all? { |t| tables.include?(t) }
  end

  # set tables
  tables = Set.new(queries.reject(&:missing_tables).flat_map(&:tables))

  # must come after missing tables set
  if @include_tables
    include_set = Set.new(@include_tables)
    tables.keep_if { |t| include_set.include?(t) || include_set.include?(t.split(".")[-1]) }
  end

  if @exclude_tables.any?
    exclude_set = Set.new(@exclude_tables)
    tables.delete_if { |t| exclude_set.include?(t) || exclude_set.include?(t.split(".")[-1]) }
  end

  # remove system tables
  tables.delete_if { |t| t.start_with?("information_schema.") || t.start_with?("pg_catalog.") }

  queries.each do |query|
    query.candidate_tables = !query.missing_tables && query.tables.any? { |t| tables.include?(t) }
  end

  # analyze tables if needed
  analyze_tables(tables) if tables.any? && (@analyze || @log_level == "debug2")

  # create hypothetical indexes and explain queries
  candidates = tables.any? ? create_hypothetical_indexes(queries.select(&:candidate_tables)) : {}

  # see if new indexes were used and meet bar
  new_indexes = determine_indexes(queries, candidates, tables)

  # display and create new indexes
  show_and_create_indexes(new_indexes, queries)
end

#process_stat_statements ⇒ Object 



25
26
27
28
29
 
# File 'lib/dexter/indexer.rb', line 25

def process_stat_statements
  queries = stat_statements.map { |q| Query.new(q) }.sort_by(&:fingerprint).group_by(&:fingerprint).map { |_, v| v.first }
  log "Processing #{queries.size} new query fingerprints"
  process_queries(queries)
end

#stat_activity ⇒ Object 



31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
 
# File 'lib/dexter/indexer.rb', line 31

def stat_activity
  execute <<~SQL
    SELECT
      pid || ':' || COALESCE(query_start, xact_start) AS id,
      query,
      EXTRACT(EPOCH FROM NOW() - COALESCE(query_start, xact_start)) * 1000.0 AS duration_ms
    FROM
      pg_stat_activity
    WHERE
      datname = current_database()
      AND state = 'active'
      AND pid != pg_backend_pid()
    ORDER BY
      1
  SQL
end

#stderr_activity ⇒ Object

works with file_fdw: www.postgresql.org/docs/current/file-fdw.html log_fdw: docs.aws.amazon.com/AmazonRDS/latest/UserGuide/Appendix.PostgreSQL.CommonDBATasks.Extensions.foreign-data-wrappers.html



68
69
70
71
72
73
74
75
76
 
# File 'lib/dexter/indexer.rb', line 68

def stderr_activity
  query = <<~SQL
    SELECT
      log_entry
    FROM
      #{conn.quote_ident(@log_table)}
  SQL
  execute(query)
end