fluentd-plugin-kinesis-intuit 2.1.1

data/gemfiles/Gemfile.fluentd-0.14.10

@@ -0,0 +1,20 @@
+#
+# Copyright 2014-2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in fluent-plugin-kinesis.gemspec
+gemspec path: ".."
+
+gem "fluentd", "0.14.10"

data/gemfiles/Gemfile.td-agent-3.2.0

@@ -0,0 +1,31 @@
+#
+# Copyright 2014-2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in fluent-plugin-kinesis.gemspec
+gemspec path: ".."
+
+# Specify related gems for td-agent v3.2.0
+# https://github.com/treasure-data/omnibus-td-agent/blob/v3.2.0/config/projects/td-agent3.rb#L27
+gem "fluentd", "1.2.2"
+# https://github.com/treasure-data/omnibus-td-agent/blob/v3.2.0/plugin_gems.rb#L16-L23
+gem "jmespath", "1.4.0"
+gem "aws-partitions", "1.87.0"
+gem "aws-sigv4", "1.0.2"
+gem "aws-sdk-core", "3.21.2"
+gem "aws-sdk-kms", "1.5.0"
+gem "aws-sdk-sqs", "1.3.0"
+gem "aws-sdk-s3", "1.13.0"
+gem "fluent-plugin-s3", "1.1.3"

data/lib/fluent/plugin/kinesis.rb

@@ -0,0 +1,146 @@
+#
+# Copyright 2014-2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+
+require 'fluent/plugin/output'
+require 'fluent/plugin/kinesis_helper/client'
+require 'fluent/plugin/kinesis_helper/api'
+require 'zlib'
+
+module Fluent
+  module Plugin
+    class KinesisOutput < Fluent::Plugin::Output
+      include Fluent::MessagePackFactory::Mixin
+      include KinesisHelper::Client
+      include KinesisHelper::API
+
+      class SkipRecordError < ::StandardError
+        def initialize(message, record)
+          super message
+          @record_message = if record.is_a? Array
+                              record.reverse.map(&:to_s).join(', ')
+                            else
+                              record.to_s
+                            end
+        end
+
+        def to_s
+          super + ": " + @record_message
+        end
+      end
+      class KeyNotFoundError < SkipRecordError
+        def initialize(key, record)
+          super "Key '#{key}' doesn't exist", record
+        end
+      end
+      class ExceedMaxRecordSizeError < SkipRecordError
+        def initialize(size, record)
+          super "Record size limit exceeded in #{size/1024} KB", record
+        end
+      end
+      class InvalidRecordError < SkipRecordError
+        def initialize(record)
+          super "Invalid type of record", record
+        end
+      end
+
+      config_param :data_key,              :string,  default: nil
+      config_param :log_truncate_max_size, :integer, default: 1024
+      config_param :compression,           :string,  default: nil
+      config_section :format do
+        config_set_default :@type, 'json'
+      end
+      config_section :inject do
+        config_set_default :time_type, 'string'
+        config_set_default :time_format, '%Y-%m-%dT%H:%M:%S.%N%z'
+      end
+
+      config_param :debug, :bool, default: false
+
+      helpers :formatter, :inject
+
+      def configure(conf)
+        super
+        @data_formatter = data_formatter_create(conf)
+      end
+
+      def multi_workers_ready?
+        true
+      end
+
+      private
+
+      def data_formatter_create(conf)
+        formatter = formatter_create
+        compressor = compressor_create
+        if @data_key.nil?
+          ->(tag, time, record) {
+            record = inject_values_to_record(tag, time, record)
+            compressor.call(formatter.format(tag, time, record).chomp.b)
+          }
+        else
+          ->(tag, time, record) {
+            raise InvalidRecordError, record unless record.is_a? Hash
+            raise KeyNotFoundError.new(@data_key, record) if record[@data_key].nil?
+            compressor.call(record[@data_key].to_s.b)
+          }
+        end
+      end
+
+      def compressor_create
+        case @compression
+        when "zlib"
+          ->(data) { Zlib::Deflate.deflate(data) }
+        else
+          ->(data) { data }
+        end
+      end
+
+      def format_for_api(&block)
+        converted = block.call
+        size = size_of_values(converted)
+        if size > @max_record_size
+          raise ExceedMaxRecordSizeError.new(size, converted)
+        end
+        converted.to_msgpack
+      rescue SkipRecordError => e
+        log.error(truncate e)
+        ''
+      end
+
+      def write_records_batch(chunk, &block)
+        unique_id = chunk.dump_unique_id_hex(chunk.unique_id)
+        chunk.open do |io|
+          records = msgpack_unpacker(io).to_enum
+          split_to_batches(records) do |batch, size|
+            log.debug(sprintf "Write chunk %s / %3d records / %4d KB", unique_id, batch.size, size/1024)
+            batch_request_with_retry(batch, &block)
+            log.debug("Finish writing chunk")
+          end
+        end
+      end
+
+      def request_type
+        self.class::RequestType
+      end
+
+      def truncate(msg)
+        if @log_truncate_max_size == 0 or (msg.to_s.size <= @log_truncate_max_size)
+          msg.to_s
+        else
+          msg.to_s[0...@log_truncate_max_size]
+        end
+      end
+    end
+  end
+end

data/lib/fluent/plugin/kinesis_helper/aggregator.rb

@@ -0,0 +1,101 @@
+#
+# Copyright 2014-2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+
+require 'fluent/configurable'
+require 'google/protobuf'
+
+Google::Protobuf::DescriptorPool.generated_pool.build do
+  add_message "AggregatedRecord" do
+    repeated :partition_key_table, :string, 1
+    repeated :explicit_hash_key_table, :string, 2
+    repeated :records, :message, 3, "Record"
+  end
+  add_message "Tag" do
+    optional :key, :string, 1
+    optional :value, :string, 2
+  end
+  add_message "Record" do
+    optional :partition_key_index, :uint64, 1
+    optional :explicit_hash_key_index, :uint64, 2
+    optional :data, :bytes, 3
+    repeated :tags, :message, 4, "Tag"
+  end
+end
+
+module Fluent
+  module Plugin
+    module KinesisHelper
+      class Aggregator
+        AggregatedRecord = Google::Protobuf::DescriptorPool.generated_pool.lookup("AggregatedRecord").msgclass
+        Tag = Google::Protobuf::DescriptorPool.generated_pool.lookup("Tag").msgclass
+        Record = Google::Protobuf::DescriptorPool.generated_pool.lookup("Record").msgclass
+
+        class InvalidEncodingError < ::StandardError; end
+
+        MagicNumber = ['F3899AC2'].pack('H*')
+
+        def aggregate(records, partition_key)
+          message = AggregatedRecord.encode(AggregatedRecord.new(
+            partition_key_table: ['a', partition_key],
+            records: records.map{|data|
+              Record.new(partition_key_index: 1, data: data)
+            },
+          ))
+          [MagicNumber, message, Digest::MD5.digest(message)].pack("A4A*A16")
+        end
+
+        def deaggregate(encoded)
+          unless aggregated?(encoded)
+            raise InvalidEncodingError, "Invalid MagicNumber #{encoded[0..3]}}"
+          end
+          message, digest = encoded[4..encoded.length-17], encoded[encoded.length-16..-1]
+          if Digest::MD5.digest(message) != digest
+            raise InvalidEncodingError, "Digest mismatch #{digest}"
+          end
+          decoded = AggregatedRecord.decode(message)
+          records = decoded.records.map(&:data)
+          partition_key = decoded.partition_key_table[1]
+          [records, partition_key]
+        end
+
+        def aggregated?(encoded)
+          encoded[0..3] == MagicNumber
+        end
+
+        def aggregated_size_offset(partition_key)
+          data = 'd'
+          encoded = aggregate([record(data)], partition_key)
+          finalize(encoded).size - data.size
+        end
+
+        module Mixin
+          AggregateOffset = 25
+          RecordOffset = 10
+
+          module Params
+            include Fluent::Configurable
+          end
+
+          def self.included(mod)
+            mod.include Params
+          end
+
+          def aggregator
+            @aggregator ||= Aggregator.new
+          end
+        end
+      end
+    end
+  end
+end

data/lib/fluent/plugin/kinesis_helper/api.rb

@@ -0,0 +1,198 @@
+#
+# Copyright 2014-2017 Amazon.com, Inc. or its affiliates. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"). You
+# may not use this file except in compliance with the License. A copy of
+# the License is located at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# or in the "license" file accompanying this file. This file is
+# distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF
+# ANY KIND, either express or implied. See the License for the specific
+# language governing permissions and limitations under the License.
+
+require 'fluent_plugin_kinesis/version'
+require 'fluent/configurable'
+
+module Fluent
+  module Plugin
+    module KinesisHelper
+      module API
+        MaxRecordSize = 1024 * 1024 # 1 MB
+
+        module APIParams
+          include Fluent::Configurable
+          config_param :max_record_size, :integer, default: MaxRecordSize
+        end
+
+        def self.included(mod)
+          mod.include APIParams
+        end
+
+        def configure(conf)
+          super
+          if @max_record_size > MaxRecordSize
+            raise ConfigError, "max_record_size can't be grater than #{MaxRecordSize/1024} KB."
+          end
+        end
+
+        module BatchRequest
+          module BatchRequestParams
+            include Fluent::Configurable
+            config_param :retries_on_batch_request, :integer, default: 8
+            config_param :reset_backoff_if_success, :bool,    default: true
+            config_param :batch_request_max_count,  :integer, default: nil
+            config_param :batch_request_max_size,   :integer, default: nil
+          end
+
+          def self.included(mod)
+            mod.include BatchRequestParams
+          end
+
+          def configure(conf)
+            super
+            if @batch_request_max_count.nil?
+              @batch_request_max_count = self.class::BatchRequestLimitCount
+            elsif @batch_request_max_count > self.class::BatchRequestLimitCount
+              raise ConfigError, "batch_request_max_count can't be grater than #{self.class::BatchRequestLimitCount}."
+            end
+            if @batch_request_max_size.nil?
+              @batch_request_max_size = self.class::BatchRequestLimitSize
+            elsif @batch_request_max_size > self.class::BatchRequestLimitSize
+              raise ConfigError, "batch_request_max_size can't be grater than #{self.class::BatchRequestLimitSize}."
+            end
+          end
+
+          def size_of_values(record)
+            record.compact.map(&:size).inject(:+) || 0
+          end
+
+          private
+
+          def split_to_batches(records, &block)
+            batch = []
+            size = 0
+            records.each do |record|
+              record_size = size_of_values(record)
+              if batch.size+1 > @batch_request_max_count or size+record_size > @batch_request_max_size
+                yield(batch, size)
+                batch = []
+                size = 0
+        end
+              batch << record
+              size += record_size
+            end
+            yield(batch, size) if batch.size > 0
+          end
+
+          def batch_request_with_retry(batch, retry_count=0, backoff: nil, &block)
+            backoff ||= Backoff.new
+            res = yield(batch)
+            if failed_count(res) > 0
+              failed_records = collect_failed_records(batch, res)
+              if retry_count < @retries_on_batch_request
+                backoff.reset if @reset_backoff_if_success and any_records_shipped?(res)
+                wait_second = backoff.next
+                msg = 'Retrying to request batch. Retry count: %3d, Retry records: %3d, Wait seconds %3.2f' % [retry_count+1, failed_records.size, wait_second]
+                log.warn(truncate msg)
+                # TODO: sleep() doesn't wait the given seconds sometime.
+                # The root cause is unknown so far, so I'd like to add debug print only. It should be fixed in the future.
+                log.debug("#{Thread.current.object_id} sleep start")
+                sleep(wait_second)
+                log.debug("#{Thread.current.object_id} sleep finish")
+                batch_request_with_retry(retry_records(failed_records), retry_count+1, backoff: backoff, &block)
+              else
+                give_up_retries(failed_records)
+              end
+            end
+          end
+
+          def any_records_shipped?(res)
+            results(res).size > failed_count(res)
+          end
+
+          def collect_failed_records(records, res)
+            failed_records = []
+            results(res).each_with_index do |record, index|
+              next unless record[:error_code]
+              original = case request_type
+                         when :streams, :firehose; records[index]
+                         when :streams_aggregated; records
+                         end
+              failed_records.push(
+                original:      original,
+                error_code:    record[:error_code],
+                error_message: record[:error_message]
+              )
+            end
+            failed_records
+          end
+
+          def retry_records(failed_records)
+            case request_type
+            when :streams, :firehose
+              failed_records.map{|r| r[:original] }
+            when :streams_aggregated
+              failed_records.first[:original]
+            end
+          end
+
+          def failed_count(res)
+            failed_field = case request_type
+                           when :streams;            :failed_record_count
+                           when :streams_aggregated; :failed_record_count
+                           when :firehose;           :failed_put_count
+                           end
+            res[failed_field]
+          end
+
+          def results(res)
+            result_field = case request_type
+                           when :streams;            :records
+                           when :streams_aggregated; :records
+                           when :firehose;           :request_responses
+                           end
+            res[result_field]
+          end
+
+          def give_up_retries(failed_records)
+            failed_records.each {|record|
+              log.error(truncate 'Could not put record, Error: %s/%s, Record: %s' % [
+                record[:error_code],
+                record[:error_message],
+                record[:original]
+              ])
+            }
+          end
+
+          class Backoff
+            def initialize
+              @count = 0
+            end
+
+            def next
+              value = calc(@count)
+              @count += 1
+              value
+            end
+
+            def reset
+              @count = 0
+            end
+
+            private
+
+            def calc(count)
+              (2 ** count) * scaling_factor
+            end
+
+            def scaling_factor
+              0.3 + (0.5-rand) * 0.1
+            end
+          end
+        end
+      end
+    end
+  end
+end