Class: Fluent::GrepCounterOutput

Inherits:
Output
  • Object
show all
Defined in:
lib/fluent/plugin/out_grepcounter.rb

Constant Summary collapse

REGEXP_MAX_NUM = 
20

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize ⇒ GrepCounterOutput

Returns a new instance of GrepCounterOutput.



7
8
9
10
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 7

def initialize
  super
  require 'pathname'
end

Instance Attribute Details

#counts ⇒ Object

Returns the value of attribute counts.



34
35
36
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 34

def counts
  @counts
end

#last_checked ⇒ Object

Returns the value of attribute last_checked.



38
39
40
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 38

def last_checked
  @last_checked
end

#matches ⇒ Object

Returns the value of attribute matches.



35
36
37
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 35

def matches
  @matches
end

#saved_at ⇒ Object

Returns the value of attribute saved_at.



37
38
39
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 37

def saved_at
  @saved_at
end

#saved_duration ⇒ Object

Returns the value of attribute saved_duration.



36
37
38
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 36

def saved_duration
  @saved_duration
end

Instance Method Details

#configure(conf) ⇒ Object 



40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
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
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 40

def configure(conf)
  super

  if @input_key
    @regexp = Regexp.compile(@regexp) if @regexp
    @exclude = Regexp.compile(@exclude) if @exclude
  end

  @regexps = {}
  (1..REGEXP_MAX_NUM).each do |i|
    next unless conf["regexp#{i}"]
    key, regexp = conf["regexp#{i}"].split(/ /, 2)
    raise Fluent::ConfigError, "regexp#{i} does not contain 2 parameters" unless regexp
    raise Fluent::ConfigError, "regexp#{i} contains a duplicated key, #{key}" if @regexps[key]
    @regexps[key] = Regexp.compile(regexp)
  end

  @excludes = {}
  (1..REGEXP_MAX_NUM).each do |i|
    next unless conf["exclude#{i}"]
    key, exclude = conf["exclude#{i}"].split(/ /, 2)
    raise Fluent::ConfigError, "exclude#{i} does not contain 2 parameters" unless exclude
    raise Fluent::ConfigError, "exclude#{i} contains a duplicated key, #{key}" if @excludes[key]
    @excludes[key] = Regexp.compile(exclude)
  end

  if @input_key and (!@regexps.empty? or !@excludes.empty?)
    raise Fluent::ConfigError, "Classic style `input_key`, and new style `regexpN`, `excludeN` can not be used together"
  end

  # to support obsolete options
  @tag ||= @output_tag
  @delimiter ||= @output_with_joined_delimiter

  # to support obsolete `threshold` and `comparator` options
  if @threshold.nil? and @less_than.nil? and @less_equal.nil? and @greater_than.nil? and @greater_equal.nil?
    @threshold = 1
  end
  unless %w[>= <=].include?(@comparator)
    raise Fluent::ConfigError, "grepcounter: comparator allows >=, <="
  end
  if @threshold
    case @comparator
    when '>='
      @greater_equal = @threshold
    else
      @less_equal = @threshold
    end
  end

  if @tag.nil? and @add_tag_prefix.nil? and @remove_tag_prefix.nil?
    @add_tag_prefix = 'count' # not ConfigError to support lower version compatibility
  end
  @tag_prefix = "#{@add_tag_prefix}." if @add_tag_prefix
  @tag_prefix_match = "#{@remove_tag_prefix}." if @remove_tag_prefix
  @tag_proc =
    if @tag
      Proc.new {|tag| @tag }
    elsif @tag_prefix and @tag_prefix_match
      Proc.new {|tag| "#{@tag_prefix}#{lstrip(tag, @tag_prefix_match)}" }
    elsif @tag_prefix_match
      Proc.new {|tag| lstrip(tag, @tag_prefix_match) }
    elsif @tag_prefix
      Proc.new {|tag| "#{@tag_prefix}#{tag}" }
    else
      Proc.new {|tag| tag }
    end

  case @aggregate
  when 'all'
    raise Fluent::ConfigError, "grepcounter: `tag` must be specified with aggregate all" if @tag.nil?
  when 'tag'
    # raise Fluent::ConfigError, "grepcounter: add_tag_prefix must be specified with aggregate tag" if @add_tag_prefix.nil?
  else
    raise Fluent::ConfigError, "grepcounter: aggregate allows tag/all"
  end

  if @store_file
    f = Pathname.new(@store_file)
    if (f.exist? && !f.writable_real?) || (!f.exist? && !f.parent.writable_real?)
      raise Fluent::ConfigError, "#{@store_file} is not writable"
    end
  end

  @matches = {}
  @counts  = {}
  @mutex = Mutex.new
end

#emit(tag, es, chain) ⇒ Object

Called when new line comes. This method actually does not emit



143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 143

def emit(tag, es, chain)
  count = 0; matches = []
  # filter out and insert
  es.each do |time,record|
    catch(:break_loop) do
      if key = @input_key
        value = record[key].to_s
        throw :break_loop if @regexp and !match(@regexp, value)
        throw :break_loop if @exclude and match(@exclude, value)
        matches << value # old style stores as an array of values
      else
        @regexps.each do |key, regexp|
          throw :break_loop unless match(regexp, record[key].to_s)
        end
        @excludes.each do |key, exclude|
          throw :break_loop if match(exclude, record[key].to_s)
        end
        matches << record # new style stores as an array of hashes, but how to utilize it?
      end
      count += 1
    end
  end
  # thread safe merge
  @counts[tag] ||= 0
  @matches[tag] ||= []
  @mutex.synchronize do
    @counts[tag] += count
    @matches[tag] += matches
  end

  chain.next
rescue => e
  $log.warn "grepcounter: #{e.class} #{e.message} #{e.backtrace.first}"
end

#flush_emit(step) ⇒ Object

This method is the real one to emit



197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 197

def flush_emit(step)
  time = Fluent::Engine.now
  flushed_counts, flushed_matches, @counts, @matches = @counts, @matches, {}, {}

  if @aggregate == 'all'
    count = 0; matches = []
    flushed_counts.keys.each do |tag|
      count += flushed_counts[tag]
      matches += flushed_matches[tag]
    end
    output = generate_output(count, matches)
    Fluent::Engine.emit(@tag, time, output) if output
  else
    flushed_counts.keys.each do |tag|
      count = flushed_counts[tag]
      matches = flushed_matches[tag]
      output = generate_output(count, matches, tag)
      if output
        emit_tag = @tag_proc.call(tag)
        Fluent::Engine.emit(emit_tag, time, output)
      end
    end
  end
end

#generate_output(count, matches, tag = nil) ⇒ Object 



222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 222

def generate_output(count, matches, tag = nil)
  return nil if count.nil?
  return nil if count == 0 # ignore 0 because standby nodes receive no message usually
  return nil if @less_than     and @less_than   <= count
  return nil if @less_equal    and @less_equal  <  count
  return nil if @greater_than  and count <= @greater_than
  return nil if @greater_equal and count <  @greater_equal
  output = {}
  output['count'] = count
  if @input_key
    output['message'] = @delimiter ? matches.join(@delimiter) : matches
  else
    # I will think of good format later...
    output['message'] = @delimiter ? matches.map{|hash| hash.to_hash}.join(@delimiter) : matches
  end
  if tag
    output['input_tag'] = tag
    output['input_tag_last'] = tag.split('.').last
  end
  output
end

#load_status(file_path, count_interval) ⇒ Object

Load internal status from a file

Parameters:

  • file_path (String)
  • count_interval (Interger) 


293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 293

def load_status(file_path, count_interval)
  return unless (f = Pathname.new(file_path)).exist?
  begin
    f.open('rb') do |f|
      stored = Marshal.load(f)
      if stored[:regexp] == @regexp and
        stored[:exclude] == @exclude and
        stored[:input_key]  == @input_key

        if Fluent::Engine.now <= stored[:saved_at] + count_interval
          @counts = stored[:counts]
          @matches = stored[:matches]
          @saved_at = stored[:saved_at]
          @saved_duration = stored[:saved_duration]

          # skip the saved duration to continue counting
          @last_checked = Fluent::Engine.now - @saved_duration
        else
          $log.warn "out_grepcounter: stored data is outdated. ignore stored data"
        end
      else
        $log.warn "out_grepcounter: configuration param was changed. ignore stored data"
      end
    end
  rescue => e
    $log.warn "out_grepcounter: Can't load store_file #{e.class} #{e.message}"
  end
end

#lstrip(string, substring) ⇒ Object 



244
245
246
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 244

def lstrip(string, substring)
  string.index(substring) == 0 ? string[substring.size..-1] : string
end

#match(regexp, string) ⇒ Object 



248
249
250
251
252
253
254
255
256
257
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 248

def match(regexp, string)
  begin
    return regexp.match(string)
  rescue ArgumentError => e
    raise e unless e.message.index("invalid byte sequence in") == 0
    string = replace_invalid_byte(string)
    retry
  end
  return true
end

#replace_invalid_byte(string) ⇒ Object 



259
260
261
262
263
264
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 259

def replace_invalid_byte(string)
  replace_options = { invalid: :replace, undef: :replace, replace: '?' }
  original_encoding = string.encoding
  temporal_encoding = (original_encoding == Encoding::UTF_8 ? Encoding::UTF_16BE : Encoding::UTF_8)
  string.encode(temporal_encoding, original_encoding, replace_options).encode(original_encoding)
end

#save_status(file_path) ⇒ Object

Store internal status into a file

Parameters:

  • file_path (String) 


269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 269

def save_status(file_path)
  begin
    Pathname.new(file_path).open('wb') do |f|
      @saved_at = Fluent::Engine.now
      @saved_duration = @saved_at - @last_checked
      Marshal.dump({
        :counts           => @counts,
        :matches          => @matches,
        :saved_at         => @saved_at,
        :saved_duration   => @saved_duration,
        :regexp           => @regexp,
        :exclude          => @exclude,
        :input_key        => @input_key,
      }, f)
    end
  rescue => e
    $log.warn "out_grepcounter: Can't write store_file #{e.class} #{e.message}"
  end
end

#shutdown ⇒ Object 



135
136
137
138
139
140
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 135

def shutdown
  super
  @watcher.terminate
  @watcher.join
  save_status(@store_file) if @store_file
end

#start ⇒ Object 



129
130
131
132
133
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 129

def start
  super
  load_status(@store_file, @count_interval) if @store_file
  @watcher = Thread.new(&method(:watcher))
end

#watcher ⇒ Object

thread callback



179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
 
# File 'lib/fluent/plugin/out_grepcounter.rb', line 179

def watcher
  # instance variable, and public accessable, for test
  @last_checked ||= Fluent::Engine.now
  while true
    sleep 0.5
    begin
      if Fluent::Engine.now - @last_checked >= @count_interval
        now = Fluent::Engine.now
        flush_emit(now - @last_checked)
        @last_checked = now
      end
    rescue => e
      $log.warn "grepcounter: #{e.class} #{e.message} #{e.backtrace.first}"
    end
  end
end