Class: Fluent::WebHDFSOutput

Inherits:
TimeSlicedOutput
  • Object
show all
Includes:
Mixin::ConfigPlaceholders, Mixin::PlainTextFormatter
Defined in:
lib/fluent/plugin/out_webhdfs.rb

Constant Summary collapse

SUPPORTED_COMPRESS = 
['gzip']
CHUNK_ID_PLACE_HOLDER = 
'${chunk_id}'

Instance Method Summary collapse

Constructor Details

#initialize ⇒ WebHDFSOutput

Returns a new instance of WebHDFSOutput.



69
70
71
72
73
74
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 69

def initialize
  super
  require 'net/http'
  require 'time'
  require 'webhdfs'
end

Instance Method Details

#chunk_unique_id_to_str(unique_id) ⇒ Object 



209
210
211
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 209

def chunk_unique_id_to_str(unique_id)
  unique_id.unpack('C*').map{|x| x.to_s(16).rjust(2,'0')}.join('')
end

#compress_context(chunk, &block) ⇒ Object 



242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 242

def compress_context(chunk, &block)
  case @compress
  when 'gzip'
    require 'zlib'
    require 'tempfile'
    tmp = Tempfile.new("webhdfs-")
    begin
      w = Zlib::GzipWriter.new(tmp)
      chunk.write_to(w)
      w.close
      tmp.close
      tmp.open
      yield tmp
    ensure
      tmp.close(true) rescue nil
    end
  end
end

#configure(conf) ⇒ Object 



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
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 81

def configure(conf)
  if conf['path']
    if conf['path'].index('%S')
      conf['time_slice_format'] = '%Y%m%d%H%M%S'
    elsif conf['path'].index('%M')
      conf['time_slice_format'] = '%Y%m%d%H%M'
    elsif conf['path'].index('%H')
      conf['time_slice_format'] = '%Y%m%d%H'
    end
  end

  super

  if @host
    @namenode_host = @host
    @namenode_port = @port
  elsif @namenode
    unless /\A([a-zA-Z0-9][-a-zA-Z0-9.]*):(\d+)\Z/ =~ @namenode
      raise Fluent::ConfigError, "Invalid config value about namenode: '#{@namenode}', needs NAMENODE_HOST:PORT"
    end
    @namenode_host = $1
    @namenode_port = $2.to_i
  else
    raise Fluent::ConfigError, "WebHDFS host or namenode missing"
  end
  if @standby_namenode
    unless /\A([a-zA-Z0-9][-a-zA-Z0-9.]*):(\d+)\Z/ =~ @standby_namenode
      raise Fluent::ConfigError, "Invalid config value about standby namenode: '#{@standby_namenode}', needs STANDBY_NAMENODE_HOST:PORT"
    end
    if @httpfs
      raise Fluent::ConfigError, "Invalid configuration: specified to use both of standby_namenode and httpfs."
    end
    @standby_namenode_host = $1
    @standby_namenode_port = $2.to_i
  end
  unless @path.index('/') == 0
    raise Fluent::ConfigError, "Path on hdfs MUST starts with '/', but '#{@path}'"
  end

  @client = prepare_client(@namenode_host, @namenode_port, @username)
  if @standby_namenode_host
    @client_standby = prepare_client(@standby_namenode_host, @standby_namenode_port, @username)
  else
    @client_standby = nil
  end

  if not @append
    if @path.index(CHUNK_ID_PLACE_HOLDER).nil?
      raise Fluent::ConfigError, "path must contain ${chunk_id}, which is the placeholder for chunk_id, when append is set to false."
    end
  end
end

#generate_path(chunk) ⇒ Object 



227
228
229
230
231
232
233
234
235
236
237
238
239
240
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 227

def generate_path(chunk)
  hdfs_path = if @append
                path_format(chunk.key)
              else
                path_format(chunk.key).gsub(CHUNK_ID_PLACE_HOLDER, chunk_unique_id_to_str(chunk.unique_id))
              end
  if @compress
    case @compress
    when 'gzip'
      hdfs_path = "#{hdfs_path}.gz"
    end
  end
  hdfs_path
end

#is_standby_exception(e) ⇒ Object 



198
199
200
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 198

def is_standby_exception(e)
  e.is_a?(WebHDFS::IOError) && e.message.match(/org\.apache\.hadoop\.ipc\.StandbyException/)
end

#namenode_available(client) ⇒ Object 



158
159
160
161
162
163
164
165
166
167
168
169
170
171
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 158

def namenode_available(client)
  if client
    available = true
    begin
      client.list('/')
    rescue => e
      log.warn "webhdfs check request failed. (namenode: #{client.host}:#{client.port}, error: #{e.message})"
      available = false
    end
    available
  else
    false
  end
end

#namenode_failover ⇒ Object 



202
203
204
205
206
207
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 202

def namenode_failover
  if @standby_namenode
    @client, @client_standby = @client_standby, @client
    log.warn "Namenode failovered, now using #{@client.host}:#{@client.port}."
  end
end

#path_format(chunk_key) ⇒ Object 



194
195
196
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 194

def path_format(chunk_key)
  Time.strptime(chunk_key, @time_slice_format).strftime(@path)
end

#prepare_client(host, port, username) ⇒ Object 



134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 134

def prepare_client(host, port, username)
  client = WebHDFS::Client.new(host, port, username)
  if @httpfs
    client.httpfs_mode = true
  end
  client.open_timeout = @open_timeout
  client.read_timeout = @read_timeout
  if @retry_known_errors
    client.retry_known_errors = true
    client.retry_interval = @retry_interval if @retry_interval
    client.retry_times = @retry_times if @retry_times
  end
  if @ssl
    client.ssl = true
    client.ssl_ca_file = @ssl_ca_file if @ssl_ca_file
    client.ssl_verify_mode = @ssl_verify_mode if @ssl_verify_mode
  end
  if @kerberos
    client.kerberos = true
  end

  client
end

#send_data(path, data) ⇒ Object

TODO check conflictions



215
216
217
218
219
220
221
222
223
224
225
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 215

def send_data(path, data)
  if @append
    begin
      @client.append(path, data)
    rescue WebHDFS::FileNotFoundError
      @client.create(path, data)
    end
  else
    @client.create(path, data, {'overwrite' => 'true'})
  end
end

#shutdown ⇒ Object 



190
191
192
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 190

def shutdown
  super
end

#start ⇒ Object 



173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 173

def start
  super

  if namenode_available(@client)
    log.info "webhdfs connection confirmed: #{@namenode_host}:#{@namenode_port}"
    return
  end
  if @client_standby && namenode_available(@client_standby)
    log.info "webhdfs connection confirmed: #{@standby_namenode_host}:#{@standby_namenode_port}"
    return
  end

  unless @ignore_start_check_error
    raise RuntimeError, "webhdfs is not available now."
  end
end

#write(chunk) ⇒ Object 



261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
 
# File 'lib/fluent/plugin/out_webhdfs.rb', line 261

def write(chunk)
  hdfs_path = generate_path(chunk)

  failovered = false
  begin
    if @compress
      compress_context(chunk) do |data|
        send_data(hdfs_path, data)
      end
    else
      send_data(hdfs_path, chunk.read)
    end
  rescue => e
    log.warn "failed to communicate hdfs cluster, path: #{hdfs_path}"

    raise e if !@client_standby || failovered

    if is_standby_exception(e) && namenode_available(@client_standby)
      log.warn "Seems the connected host status is not active (maybe due to failovers). Gonna try another namenode immediately."
      namenode_failover
      failovered = true
      retry
    end
    if @num_errors && ((@num_errors + 1) >= @failures_before_use_standby) && namenode_available(@client_standby)
      log.warn "Too many failures. Try to use the standby namenode instead."
      namenode_failover
      failovered = true
      retry
    end
    raise e
  end
  hdfs_path
end