Class: LogStash::Inputs::S3

Inherits:
Base
  • Object
show all
Includes:
PluginMixins::AwsConfig::V2
Defined in:
lib/logstash/inputs/s3.rb

Overview

Stream events from files from a S3 bucket.

Each line from each file generates an event. Files ending in ‘.gz` are handled as gzip’ed files.

Defined Under Namespace

Modules: SinceDB

Constant Summary collapse

CUTOFF_SECOND = 
3

Instance Method Summary collapse

Constructor Details

#initialize(*params) ⇒ S3

Returns a new instance of S3.



91
92
93
94
95
 
# File 'lib/logstash/inputs/s3.rb', line 91

def initialize(*params)
  super
  @cloudfront_fields_key = ecs_select[disabled: 'cloudfront_fields', v1: '[@metadata][s3][cloudfront][fields]']
  @cloudfront_version_key = ecs_select[disabled: 'cloudfront_version', v1: '[@metadata][s3][cloudfront][version]']
end

Instance Method Details

#backup_to_bucket(object) ⇒ Object

def fetch_new_files



166
167
168
169
170
171
172
173
174
 
# File 'lib/logstash/inputs/s3.rb', line 166

def backup_to_bucket(object)
  unless @backup_to_bucket.nil?
    backup_key = "#{@backup_add_prefix}#{object.key}"
    @backup_bucket.object(backup_key).copy_from(:copy_source => "#{object.bucket_name}/#{object.key}")
    if @delete
      object.delete()
    end
  end
end

#backup_to_dir(filename) ⇒ Object 



176
177
178
179
180
 
# File 'lib/logstash/inputs/s3.rb', line 176

def backup_to_dir(filename)
  unless @backup_to_dir.nil?
    FileUtils.cp(filename, @backup_to_dir)
  end
end

#list_new_files ⇒ Object

def run



135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
 
# File 'lib/logstash/inputs/s3.rb', line 135

def list_new_files
  objects = []
  found = false
  current_time = Time.now
  sincedb_time = sincedb.read
  begin
    @s3bucket.objects(:prefix => @prefix).each do |log|
      found = true
      @logger.debug('Found key', :key => log.key)
      if ignore_filename?(log.key)
        @logger.debug('Ignoring', :key => log.key)
      elsif log.content_length <= 0
        @logger.debug('Object Zero Length', :key => log.key)
      elsif log.last_modified <= sincedb_time
        @logger.debug('Object Not Modified', :key => log.key)
      elsif log.last_modified > (current_time - CUTOFF_SECOND).utc # file modified within last two seconds will be processed in next cycle
        @logger.debug('Object Modified After Cutoff Time', :key => log.key)
      elsif (log.storage_class == 'GLACIER' || log.storage_class == 'DEEP_ARCHIVE') && !file_restored?(log.object)
        @logger.debug('Object Archived to Glacier', :key => log.key)
      else
        objects << log
        @logger.debug("Added to objects[]", :key => log.key, :length => objects.length)
      end
    end
    @logger.info('No files found in bucket', :prefix => prefix) unless found
  rescue Aws::Errors::ServiceError => e
    @logger.error("Unable to list objects in bucket", :exception => e.class, :message => e.message, :backtrace => e.backtrace, :prefix => prefix)
  end
  objects.sort_by { |log| log.last_modified }
end

#process_files(queue) ⇒ Object 



182
183
184
185
186
187
188
189
190
191
192
 
# File 'lib/logstash/inputs/s3.rb', line 182

def process_files(queue)
  objects = list_new_files

  objects.each do |log|
    if stop?
      break
    else
      process_log(queue, log)
    end
  end
end

#register ⇒ Object 



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
 
# File 'lib/logstash/inputs/s3.rb', line 97

def register
  require "fileutils"
  require "digest/md5"

  @logger.info("Registering", :bucket => @bucket, :region => @region)

  s3 = get_s3object

  @s3bucket = s3.bucket(@bucket)

  unless @backup_to_bucket.nil?
    @backup_bucket = s3.bucket(@backup_to_bucket)
    begin
      s3.client.head_bucket({ :bucket => @backup_to_bucket})
    rescue Aws::S3::Errors::NoSuchBucket
      s3.create_bucket({ :bucket => @backup_to_bucket})
    end
  end

  unless @backup_to_dir.nil?
    Dir.mkdir(@backup_to_dir, 0700) unless File.exists?(@backup_to_dir)
  end

  FileUtils.mkdir_p(@temporary_directory) unless Dir.exist?(@temporary_directory)

  if !@watch_for_new_files && original_params.include?('interval')
    logger.warn("`watch_for_new_files` has been disabled; `interval` directive will be ignored.")
  end
end

#run(queue) ⇒ Object 



127
128
129
130
131
132
133
 
# File 'lib/logstash/inputs/s3.rb', line 127

def run(queue)
  @current_thread = Thread.current
  Stud.interval(@interval) do
    process_files(queue)
    stop unless @watch_for_new_files
  end
end

#stop ⇒ Object

def process_files



194
195
196
197
198
199
 
# File 'lib/logstash/inputs/s3.rb', line 194

def stop
  # @current_thread is initialized in the `#run` method,
  # this variable is needed because the `#stop` is a called in another thread
  # than the `#run` method and requiring us to call stop! with a explicit thread.
  Stud.stop!(@current_thread)
end