fluentd 1.13.3 → 1.16.5

data/lib/fluent/plugin/buf_file_single.rb

@@ -160,13 +160,20 @@ module Fluent
       def resume
         stage = {}
         queue = []
+        exist_broken_file = false
 
         patterns = [@path]
         patterns.unshift @additional_resume_path if @additional_resume_path
         Dir.glob(escaped_patterns(patterns)) do |path|
           next unless File.file?(path)
 
-          log.debug { "restoring buffer file: path = #{path}" }
+          if owner.respond_to?(:buffer_config) && owner.buffer_config&.flush_at_shutdown
+            # When `flush_at_shutdown` is `true`, the remaining chunk files during resuming are possibly broken
+            # since there may be a power failure or similar failure.
+            log.warn { "restoring buffer file: path = #{path}" }
+          else
+            log.debug { "restoring buffer file: path = #{path}" }
+          end
 
           m = new_metadata() # this metadata will be updated in FileSingleChunk.new
           mode = Fluent::Plugin::Buffer::FileSingleChunk.assume_chunk_state(path)
@@ -179,6 +186,7 @@ module Fluent
             chunk = Fluent::Plugin::Buffer::FileSingleChunk.new(m, path, mode, @key_in_path, compress: @compress)
             chunk.restore_size(@chunk_format) if @calc_num_records
           rescue Fluent::Plugin::Buffer::FileSingleChunk::FileChunkError => e
+            exist_broken_file = true
             handle_broken_files(path, mode, e)
             next
           end
@@ -193,6 +201,15 @@ module Fluent
 
         queue.sort_by!(&:modified_at)
 
+        # If one of the files is corrupted, other files may also be corrupted and be undetected.
+        # The time priods of each chunk are helpful to check the data.
+        if exist_broken_file
+          log.info "Since a broken chunk file was found, it is possible that other files remaining at the time of resuming were also broken. Here is the list of the files."
+          (stage.values + queue).each { |chunk|
+            log.info "  #{chunk.path}:", :created_at => chunk.created_at, :modified_at => chunk.modified_at
+          }
+        end
+
         return stage, queue
       end
 
@@ -207,8 +224,20 @@ module Fluent
       end
 
       def handle_broken_files(path, mode, e)
-        log.error "found broken chunk file during resume. Delete corresponding files:", path: path, mode: mode, err_msg: e.message
-        # After support 'backup_dir' feature, these files are moved to backup_dir instead of unlink.
+        log.error "found broken chunk file during resume.", :path => path, :mode => mode, :err_msg => e.message
+        unique_id, _ = Fluent::Plugin::Buffer::FileSingleChunk.unique_id_and_key_from_path(path)
+        backup(unique_id) { |f|
+          File.open(path, 'rb') { |chunk|
+            chunk.set_encoding(Encoding::ASCII_8BIT)
+            chunk.sync = true
+            chunk.binmode
+            IO.copy_stream(chunk, f)
+          }
+        }
+      rescue => error
+        log.error "backup failed. Delete corresponding files.", :err_msg => error.message
+      ensure
+        log.warn "disable_chunk_backup is true. #{dump_unique_id_hex(unique_id)} chunk is thrown away." if @disable_chunk_backup
         File.unlink(path) rescue nil
       end
 

data/lib/fluent/plugin/buffer/file_chunk.rb

@@ -204,7 +204,7 @@ module Fluent
           end
         end
 
-        # used only for queued v0.12 buffer path
+        # used only for queued v0.12 buffer path or broken files
         def self.unique_id_from_path(path)
           if /\.(b|q)([0-9a-f]+)\.[^\/]*\Z/n =~ path # //n switch means explicit 'ASCII-8BIT' pattern
             return $2.scan(/../).map{|x| x.to_i(16) }.pack('C*')

data/lib/fluent/plugin/buffer.rb

@@ -16,6 +16,8 @@
 
 require 'fluent/plugin/base'
 require 'fluent/plugin/owned_by_mixin'
+require 'fluent/plugin_id'
+require 'fluent/plugin_helper'
 require 'fluent/unique_id'
 require 'fluent/ext_monitor_require'
 
@@ -24,7 +26,9 @@ module Fluent
     class Buffer < Base
       include OwnedByMixin
       include UniqueId::Mixin
+      include PluginId
       include MonitorMixin
+      include PluginHelper::Mixin # for metrics
 
       class BufferError < StandardError; end
       class BufferOverflowError < BufferError; end
@@ -39,6 +43,8 @@ module Fluent
 
       configured_in :buffer
 
+      helpers_internal :metrics
+
       # TODO: system total buffer limit size in bytes by SystemConfig
 
       config_param :chunk_limit_size, :size, default: DEFAULT_CHUNK_LIMIT_SIZE
@@ -60,6 +66,9 @@ module Fluent
       desc 'Compress buffered data.'
       config_param :compress, :enum, list: [:text, :gzip], default: :text
 
+      desc 'If true, chunks are thrown away when unrecoverable error happens'
+      config_param :disable_chunk_backup, :bool, default: false
+
       Metadata = Struct.new(:timekey, :tag, :variables, :seq) do
         def initialize(timekey, tag, variables)
           super(timekey, tag, variables, 0)
@@ -153,8 +162,11 @@ module Fluent
         end
       end
 
+      # for metrics
+      attr_reader :stage_size_metrics, :stage_length_metrics, :queue_size_metrics, :queue_length_metrics
+      attr_reader :available_buffer_space_ratios_metrics, :total_queued_size_metrics
+      attr_reader :newest_timekey_metrics, :oldest_timekey_metrics
       # for tests
-      attr_accessor :stage_size, :queue_size
       attr_reader :stage, :queue, :dequeued, :queued_num
 
       def initialize
@@ -171,12 +183,35 @@ module Fluent
         @queued_num = {} # metadata => int (number of queued chunks)
         @dequeued_num = {} # metadata => int (number of dequeued chunks)
 
-        @stage_size = @queue_size = 0
+        @stage_length_metrics = nil
+        @stage_size_metrics = nil
+        @queue_length_metrics = nil
+        @queue_size_metrics = nil
+        @available_buffer_space_ratios_metrics = nil
+        @total_queued_size_metrics = nil
+        @newest_timekey_metrics = nil
+        @oldest_timekey_metrics = nil
         @timekeys = Hash.new(0)
         @enable_update_timekeys = false
         @mutex = Mutex.new
       end
 
+      def stage_size
+        @stage_size_metrics.get
+      end
+
+      def stage_size=(value)
+        @stage_size_metrics.set(value)
+      end
+
+      def queue_size
+        @queue_size_metrics.get
+      end
+
+      def queue_size=(value)
+        @queue_size_metrics.set(value)
+      end
+
       def persistent?
         false
       end
@@ -187,6 +222,28 @@ module Fluent
         unless @queue_limit_length.nil?
           @total_limit_size = @chunk_limit_size * @queue_limit_length
         end
+        @stage_length_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "stage_length",
+                                               help_text: 'Length of stage buffers', prefer_gauge: true)
+        @stage_length_metrics.set(0)
+        @stage_size_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "stage_byte_size",
+                                             help_text: 'Total size of stage buffers', prefer_gauge: true)
+        @stage_size_metrics.set(0) # Ensure zero.
+        @queue_length_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "queue_length",
+                                               help_text: 'Length of queue buffers', prefer_gauge: true)
+        @queue_length_metrics.set(0)
+        @queue_size_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "queue_byte_size",
+                                             help_text: 'Total size of queue buffers', prefer_gauge: true)
+        @queue_size_metrics.set(0) # Ensure zero.
+        @available_buffer_space_ratios_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "available_buffer_space_ratios",
+                                                                help_text: 'Ratio of available space in buffer', prefer_gauge: true)
+        @available_buffer_space_ratios_metrics.set(100) # Default is 100%.
+        @total_queued_size_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "total_queued_size",
+                                                    help_text: 'Total size of stage and queue buffers', prefer_gauge: true)
+        @total_queued_size_metrics.set(0)
+        @newest_timekey_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "newest_timekey",
+                                                 help_text: 'Newest timekey in buffer', prefer_gauge: true)
+        @oldest_timekey_metrics = metrics_create(namespace: "fluentd", subsystem: "buffer", name: "oldest_timekey",
+                                                 help_text: 'Oldest timekey in buffer', prefer_gauge: true)
       end
 
       def enable_update_timekeys
@@ -198,15 +255,15 @@ module Fluent
 
         @stage, @queue = resume
         @stage.each_pair do |metadata, chunk|
-          @stage_size += chunk.bytesize
+          @stage_size_metrics.add(chunk.bytesize)
         end
         @queue.each do |chunk|
           @queued_num[chunk.metadata] ||= 0
           @queued_num[chunk.metadata] += 1
-          @queue_size += chunk.bytesize
+          @queue_size_metrics.add(chunk.bytesize)
         end
         update_timekeys
-        log.debug "buffer started", instance: self.object_id, stage_size: @stage_size, queue_size: @queue_size
+        log.debug "buffer started", instance: self.object_id, stage_size: @stage_size_metrics.get, queue_size: @queue_size_metrics.get
       end
 
       def close
@@ -228,17 +285,19 @@ module Fluent
       def terminate
         super
         @dequeued = @stage = @queue = @queued_num = nil
-        @stage_size = @queue_size = 0
+        @stage_length_metrics = @stage_size_metrics = @queue_length_metrics = @queue_size_metrics = nil
+        @available_buffer_space_ratios_metrics = @total_queued_size_metrics = nil
+        @newest_timekey_metrics = @oldest_timekey_metrics = nil
         @timekeys.clear
       end
 
       def storable?
-        @total_limit_size > @stage_size + @queue_size
+        @total_limit_size > @stage_size_metrics.get + @queue_size_metrics.get
       end
 
       ## TODO: for back pressure feature
       # def used?(ratio)
-      #   @total_limit_size * ratio > @stage_size + @queue_size
+      #   @total_limit_size * ratio > @stage_size_metrics.get + @queue_size_metrics.get
       # end
 
       def resume
@@ -276,12 +335,14 @@ module Fluent
         unstaged_chunks = {} # metadata => [chunk, chunk, ...]
         chunks_to_enqueue = []
         staged_bytesizes_by_chunk = {}
+        # track internal BufferChunkOverflowError in write_step_by_step
+        buffer_chunk_overflow_errors = []
 
         begin
           # sort metadata to get lock of chunks in same order with other threads
           metadata_and_data.keys.sort.each do |metadata|
             data = metadata_and_data[metadata]
-            write_once(metadata, data, format: format, size: size) do |chunk, adding_bytesize|
+            write_once(metadata, data, format: format, size: size) do |chunk, adding_bytesize, error|
               chunk.mon_enter # add lock to prevent to be committed/rollbacked from other threads
               operated_chunks << chunk
               if chunk.staged?
@@ -296,6 +357,9 @@ module Fluent
                 unstaged_chunks[metadata] ||= []
                 unstaged_chunks[metadata] << chunk
               end
+              if error && !error.empty?
+                buffer_chunk_overflow_errors << error
+              end
             end
           end
 
@@ -344,7 +408,7 @@ module Fluent
           #
           staged_bytesizes_by_chunk.each do |chunk, bytesize|
             chunk.synchronize do
-              synchronize { @stage_size += bytesize }
+              synchronize { @stage_size_metrics.add(bytesize) }
               log.on_trace { log.trace { "chunk #{chunk.path} size_added: #{bytesize} new_size: #{chunk.bytesize}" } }
             end
           end
@@ -353,7 +417,7 @@ module Fluent
             if c.staged? && (enqueue || chunk_size_full?(c))
               m = c.metadata
               enqueue_chunk(m)
-              if unstaged_chunks[m]
+              if unstaged_chunks[m] && !unstaged_chunks[m].empty?
                 u = unstaged_chunks[m].pop
                 u.synchronize do
                   if u.unstaged? && !chunk_size_full?(u)
@@ -361,7 +425,7 @@ module Fluent
                     u.metadata.seq = 0
                     synchronize {
                       @stage[m] = u.staged!
-                      @stage_size += u.bytesize
+                      @stage_size_metrics.add(u.bytesize)
                     }
                   end
                 end
@@ -388,6 +452,10 @@ module Fluent
             end
             chunk.mon_exit rescue nil # this may raise ThreadError for chunks already committed
           end
+          unless buffer_chunk_overflow_errors.empty?
+            # Notify delayed BufferChunkOverflowError here
+            raise BufferChunkOverflowError, buffer_chunk_overflow_errors.join(", ")
+          end
         end
       end
 
@@ -428,8 +496,8 @@ module Fluent
               chunk.enqueued!
             end
             bytesize = chunk.bytesize
-            @stage_size -= bytesize
-            @queue_size += bytesize
+            @stage_size_metrics.sub(bytesize)
+            @queue_size_metrics.add(bytesize)
           end
         end
         nil
@@ -446,7 +514,7 @@ module Fluent
             @queued_num[metadata] = @queued_num.fetch(metadata, 0) + 1
             chunk.enqueued!
           end
-          @queue_size += chunk.bytesize
+          @queue_size_metrics.add(chunk.bytesize)
         end
       end
 
@@ -512,7 +580,7 @@ module Fluent
           chunk = @dequeued.delete(chunk_id)
           return false unless chunk # already purged by other thread
           @queue.unshift(chunk)
-          log.trace "chunk taken back", instance: self.object_id, chunk_id: dump_unique_id_hex(chunk_id), metadata: chunk.metadata
+          log.on_trace { log.trace "chunk taken back", instance: self.object_id, chunk_id: dump_unique_id_hex(chunk_id), metadata: chunk.metadata }
           @queued_num[chunk.metadata] += 1 # BUG if nil
           @dequeued_num[chunk.metadata] -= 1
         end
@@ -531,7 +599,7 @@ module Fluent
           begin
             bytesize = chunk.bytesize
             chunk.purge
-            @queue_size -= bytesize
+            @queue_size_metrics.sub(bytesize)
           rescue => e
             log.error "failed to purge buffer chunk", chunk_id: dump_unique_id_hex(chunk_id), error_class: e.class, error: e
             log.error_backtrace
@@ -542,7 +610,7 @@ module Fluent
             @queued_num.delete(metadata)
             @dequeued_num.delete(metadata)
           end
-          log.trace "chunk purged", instance: self.object_id, chunk_id: dump_unique_id_hex(chunk_id), metadata: metadata
+          log.on_trace { log.trace "chunk purged", instance: self.object_id, chunk_id: dump_unique_id_hex(chunk_id), metadata: metadata }
         end
 
         nil
@@ -562,7 +630,7 @@ module Fluent
               log.error_backtrace
             end
           end
-          @queue_size = 0
+          @queue_size_metrics.set(0)
         end
       end
 
@@ -680,16 +748,14 @@ module Fluent
         modified_chunks = []
         modified_metadata = metadata
         get_next_chunk = ->(){
-          c = if staged_chunk_used
-                # Staging new chunk here is bad idea:
-                # Recovering whole state including newly staged chunks is much harder than current implementation.
-                modified_metadata = modified_metadata.dup_next
-                generate_chunk(modified_metadata)
-              else
-                synchronize { @stage[modified_metadata] ||= generate_chunk(modified_metadata).staged! }
-              end
-          modified_chunks << c
-          c
+          if staged_chunk_used
+            # Staging new chunk here is bad idea:
+            # Recovering whole state including newly staged chunks is much harder than current implementation.
+            modified_metadata = modified_metadata.dup_next
+            generate_chunk(modified_metadata)
+          else
+            synchronize { @stage[modified_metadata] ||= generate_chunk(modified_metadata).staged! }
+          end
         }
 
         writing_splits_index = 0
@@ -697,60 +763,116 @@ module Fluent
 
         while writing_splits_index < splits.size
           chunk = get_next_chunk.call
-          chunk.synchronize do
-            raise ShouldRetry unless chunk.writable?
-            staged_chunk_used = true if chunk.staged?
+          errors = []
+          # The chunk must be locked until being passed to &block.
+          chunk.mon_enter
+          modified_chunks << {chunk: chunk, adding_bytesize: 0, errors: errors}
 
-            original_bytesize = chunk.bytesize
-            begin
-              while writing_splits_index < splits.size
-                split = splits[writing_splits_index]
-                if format
-                  chunk.concat(format.call(split), split.size)
-                else
-                  chunk.append(split, compress: @compress)
+          raise ShouldRetry unless chunk.writable?
+          staged_chunk_used = true if chunk.staged?
+
+          original_bytesize = committed_bytesize = chunk.bytesize
+          begin
+            while writing_splits_index < splits.size
+              split = splits[writing_splits_index]
+              formatted_split = format ? format.call(split) : nil
+
+              if split.size == 1 # Check BufferChunkOverflowError
+                determined_bytesize = nil
+                if @compress != :text
+                  determined_bytesize = nil
+                elsif formatted_split
+                  determined_bytesize = formatted_split.bytesize
+                elsif split.first.respond_to?(:bytesize)
+                  determined_bytesize = split.first.bytesize
                 end
 
-                if chunk_size_over?(chunk) # split size is larger than difference between size_full? and size_over?
-                  chunk.rollback
+                if determined_bytesize && determined_bytesize > @chunk_limit_size
+                  # It is a obvious case that BufferChunkOverflowError should be raised here.
+                  # But if it raises here, already processed 'split' or
+                  # the proceeding 'split' will be lost completely.
+                  # So it is a last resort to delay raising such a exception
+                  errors << "a #{determined_bytesize} bytes record (nth: #{writing_splits_index}) is larger than buffer chunk limit size (#{@chunk_limit_size})"
+                  writing_splits_index += 1
+                  next
+                end
 
-                  if split.size == 1 && original_bytesize == 0
-                    big_record_size = format ? format.call(split).bytesize : split.first.bytesize
-                    raise BufferChunkOverflowError, "a #{big_record_size}bytes record is larger than buffer chunk limit size"
-                  end
+                if determined_bytesize.nil? || chunk.bytesize + determined_bytesize > @chunk_limit_size
+                  # The split will (might) cause size over so keep already processed
+                  # 'split' content here (allow performance regression a bit).
+                  chunk.commit
+                  committed_bytesize = chunk.bytesize
+                end
+              end
+
+              if format
+                chunk.concat(formatted_split, split.size)
+              else
+                chunk.append(split, compress: @compress)
+              end
+              adding_bytes = chunk.bytesize - committed_bytesize
+
+              if chunk_size_over?(chunk) # split size is larger than difference between size_full? and size_over?
+                chunk.rollback
+                committed_bytesize = chunk.bytesize
 
-                  if chunk_size_full?(chunk) || split.size == 1
-                    enqueue_chunk_before_retry = true
+                if split.size == 1 # Check BufferChunkOverflowError again
+                  if adding_bytes > @chunk_limit_size
+                    errors << "concatenated/appended a #{adding_bytes} bytes record (nth: #{writing_splits_index}) is larger than buffer chunk limit size (#{@chunk_limit_size})"
+                    writing_splits_index += 1
+                    next
                   else
-                    splits_count *= 10
+                    # As already processed content is kept after rollback, then unstaged chunk should be queued.
+                    # After that, re-process current split again.
+                    # New chunk should be allocated, to do it, modify @stage and so on.
+                    synchronize { @stage.delete(modified_metadata) }
+                    staged_chunk_used = false
+                    chunk.unstaged!
+                    break
                   end
+                end
 
-                  raise ShouldRetry
+                if chunk_size_full?(chunk) || split.size == 1
+                  enqueue_chunk_before_retry = true
+                else
+                  splits_count *= 10
                 end
 
-                writing_splits_index += 1
+                raise ShouldRetry
+              end
+
+              writing_splits_index += 1
 
-                if chunk_size_full?(chunk)
-                  break
-                end
+              if chunk_size_full?(chunk)
+                break
               end
-            rescue
-              chunk.purge if chunk.unstaged? # unstaged chunk will leak unless purge it
-              raise
             end
-
-            block.call(chunk, chunk.bytesize - original_bytesize)
+          rescue
+            chunk.purge if chunk.unstaged? # unstaged chunk will leak unless purge it
+            raise
           end
+
+          modified_chunks.last[:adding_bytesize] = chunk.bytesize - original_bytesize
+        end
+        modified_chunks.each do |data|
+          block.call(data[:chunk], data[:adding_bytesize], data[:errors])
         end
       rescue ShouldRetry
-        modified_chunks.each do |mc|
-          mc.rollback rescue nil
-          if mc.unstaged?
-            mc.purge rescue nil
+        modified_chunks.each do |data|
+          chunk = data[:chunk]
+          chunk.rollback rescue nil
+          if chunk.unstaged?
+            chunk.purge rescue nil
           end
+          chunk.mon_exit rescue nil
         end
         enqueue_chunk(metadata) if enqueue_chunk_before_retry
         retry
+      ensure
+        modified_chunks.each do |data|
+          chunk = data[:chunk]
+          chunk.mon_exit
+        end
       end
 
       STATS_KEYS = [
@@ -765,28 +887,52 @@ module Fluent
       ]
 
       def statistics
-        stage_size, queue_size = @stage_size, @queue_size
+        stage_size, queue_size = @stage_size_metrics.get, @queue_size_metrics.get
         buffer_space = 1.0 - ((stage_size + queue_size * 1.0) / @total_limit_size)
+        @stage_length_metrics.set(@stage.size)
+        @queue_length_metrics.set(@queue.size)
+        @available_buffer_space_ratios_metrics.set(buffer_space * 100)
+        @total_queued_size_metrics.set(stage_size + queue_size)
         stats = {
-          'stage_length' => @stage.size,
+          'stage_length' => @stage_length_metrics.get,
           'stage_byte_size' => stage_size,
-          'queue_length' => @queue.size,
+          'queue_length' => @queue_length_metrics.get,
           'queue_byte_size' => queue_size,
-          'available_buffer_space_ratios' => (buffer_space * 100).round(1),
-          'total_queued_size' => stage_size + queue_size,
+          'available_buffer_space_ratios' => @available_buffer_space_ratios_metrics.get.round(1),
+          'total_queued_size' => @total_queued_size_metrics.get,
         }
 
         tkeys = timekeys
         if (m = tkeys.min)
-          stats['oldest_timekey'] = m
+          @oldest_timekey_metrics.set(m)
+          stats['oldest_timekey'] = @oldest_timekey_metrics.get
         end
         if (m = tkeys.max)
-          stats['newest_timekey'] = m
+          @newest_timekey_metrics.set(m)
+          stats['newest_timekey'] = @newest_timekey_metrics.get
         end
 
         { 'buffer' => stats }
       end
 
+      def backup(chunk_unique_id)
+        unique_id = dump_unique_id_hex(chunk_unique_id)
+
+        if @disable_chunk_backup
+          log.warn "disable_chunk_backup is true. #{unique_id} chunk is not backed up."
+          return
+        end
+
+        safe_owner_id = owner.plugin_id.gsub(/[ "\/\\:;|*<>?]/, '_')
+        backup_base_dir = system_config.root_dir || DEFAULT_BACKUP_DIR
+        backup_file = File.join(backup_base_dir, 'backup', "worker#{fluentd_worker_id}", safe_owner_id, "#{unique_id}.log")
+        backup_dir = File.dirname(backup_file)
+
+        log.warn "bad chunk is moved to #{backup_file}"
+        FileUtils.mkdir_p(backup_dir, mode: system_config.dir_permission || Fluent::DEFAULT_DIR_PERMISSION) unless Dir.exist?(backup_dir)
+        File.open(backup_file, 'ab', system_config.file_permission || Fluent::DEFAULT_FILE_PERMISSION) { |f| yield f }
+      end
+
       private
 
       def optimistic_queued?(metadata = nil)

data/lib/fluent/plugin/filter.rb

@@ -28,13 +28,47 @@ module Fluent
       include PluginLoggerMixin
       include PluginHelper::Mixin
 
-      helpers_internal :event_emitter
+      helpers_internal :event_emitter, :metrics
 
       attr_reader :has_filter_with_time
 
       def initialize
         super
         @has_filter_with_time = has_filter_with_time?
+        @emit_records_metrics = nil
+        @emit_size_metrics = nil
+        @counter_mutex = Mutex.new
+        @enable_size_metrics = false
+      end
+
+      def emit_records
+        @emit_records_metrics.get
+      end
+
+      def emit_size
+        @emit_size_metrics.get
+      end
+
+      def configure(conf)
+        super
+
+        @emit_records_metrics = metrics_create(namespace: "fluentd", subsystem: "filter", name: "emit_records", help_text: "Number of count emit records")
+        @emit_size_metrics = metrics_create(namespace: "fluentd", subsystem: "filter", name: "emit_size", help_text: "Total size of emit events")
+        @enable_size_metrics = !!system_config.enable_size_metrics
+      end
+
+      def statistics
+        stats = {
+          'emit_records' => @emit_records_metrics.get,
+          'emit_size' => @emit_size_metrics.get,
+        }
+
+        { 'filter' => stats }
+      end
+
+      def measure_metrics(es)
+        @emit_records_metrics.add(es.size)
+        @emit_size_metrics.add(es.to_msgpack_stream.bytesize) if @enable_size_metrics
       end
 
       def filter(tag, time, record)

data/lib/fluent/plugin/filter_record_transformer.rb

@@ -316,7 +316,7 @@ module Fluent::Plugin
         end
 
         (Object.instance_methods).each do |m|
-          undef_method m unless m.to_s =~ /^__|respond_to_missing\?|object_id|public_methods|instance_eval|method_missing|define_singleton_method|respond_to\?|new_ostruct_member|^class$/
+          undef_method m unless /^__|respond_to_missing\?|object_id|public_methods|instance_eval|method_missing|define_singleton_method|respond_to\?|new_ostruct_member|^class$/.match?(m.to_s)
         end
       end
     end

data/lib/fluent/plugin/in_forward.rb

@@ -40,7 +40,7 @@ module Fluent::Plugin
     config_param :backlog, :integer, default: nil
     # SO_LINGER 0 to send RST rather than FIN to avoid lots of connections sitting in TIME_WAIT at src
     desc 'The timeout time used to set linger option.'
-    config_param :linger_timeout, :integer, default: 0
+    config_param :linger_timeout, :integer, default: nil, deprecated: "use transport directive"
     # This option is for Cool.io's loop wait timeout to avoid loop stuck at shutdown. Almost users don't need to change this value.
     config_param :blocking_timeout, :time, default: 0.5
     desc 'Try to resolve hostname from IP addresses or not.'
@@ -430,7 +430,7 @@ module Fluent::Plugin
       end
       _ping, hostname, shared_key_salt, shared_key_hexdigest, username, password_digest = message
 
-      node = @nodes.select{|n| n[:address].include?(remote_addr) rescue false }.first
+      node = @nodes.find{|n| n[:address].include?(remote_addr) rescue false }
       if !node && !@security.allow_anonymous_source
         log.warn "Anonymous client disallowed", address: remote_addr, hostname: hostname
         return false, "anonymous source host '#{remote_addr}' denied", nil