logstash-output-influxdb 0.1.0

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    M2VjY2UwMTc3YTdlNWVlNDI3OWY1ZjZmY2I3NzgxNjg3MzQ0MGVmYQ==
+  data.tar.gz: !binary |-
+    MzkxYThmOTdkOTE0MDAyODhiMjZmM2FiM2MxMWFjYmM3ZTlmNjA4ZQ==
+SHA512:
+  metadata.gz: !binary |-
+    NmRhZTVjNDk5ZTQ3YTY5YWU5MzY5NTVkZjk0MGQwNGY1N2MxYjZmMmVjZjA0
+    ODE5ZWNhZGIxMTQxYjQwYWM2OGMzYjc5NzE3MzIzYjBkNWM1OWMwZjBiZWM4
+    Y2UxNzI5ODNlZDZhNzFjNDNlM2VlMjEyNjM5NjUwMGU0NGZmM2U=
+  data.tar.gz: !binary |-
+    YWRjMWM1MTdhNjYxYzBlM2ZjYTE3NjAwMThjNGE5MGYyNTY5M2ZhYjQzZjY2
+    MDEwZGQ4OGI4ZjRjZWEwNTRjOWI3ZmE4OTlkNjY2NWU0ODg3YmQ5OWIwNjRj
+    ODMwZjMwODAxZTQwMzE2OThmMjUzNTYxZTJhZmZmNzY0YzIwZTU=

data/.gitignore

@@ -0,0 +1,4 @@
+*.gem
+Gemfile.lock
+.bundle
+vendor

data/Gemfile

@@ -0,0 +1,4 @@
+source 'http://rubygems.org'
+gem 'rake'
+gem 'gem_publisher'
+gem 'archive-tar-minitar'

data/LICENSE

@@ -0,0 +1,13 @@
+Copyright (c) 2012-2014 Elasticsearch <http://www.elasticsearch.org>
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License 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.

data/Rakefile

@@ -0,0 +1,6 @@
+@files=[]
+
+task :default do
+  system("rake -T")
+end
+

data/lib/logstash/outputs/influxdb.rb

@@ -0,0 +1,231 @@
+# encoding: utf-8
+require "logstash/namespace"
+require "logstash/outputs/base"
+require "stud/buffer"
+
+# This output lets you output Metrics to InfluxDB
+#
+# The configuration here attempts to be as friendly as possible
+# and minimize the need for multiple definitions to write to
+# multiple series and still be efficient
+#
+# the InfluxDB API let's you do some semblance of bulk operation
+# per http call but each call is database-specific
+#
+# You can learn more about InfluxDB at <http://influxdb.org>
+class LogStash::Outputs::InfluxDB < LogStash::Outputs::Base
+  include Stud::Buffer
+
+  config_name "influxdb"
+  milestone 1
+
+  # The database to write
+  config :db, :validate => :string, :default => "stats"
+
+  # The hostname or IP address to reach your InfluxDB instance
+  config :host, :validate => :string, :required => true
+
+  # The port for InfluxDB
+  config :port, :validate => :number, :default => 8086
+
+  # The user who has access to the named database
+  config :user, :validate => :string, :default => nil, :required => true
+
+  # The password for the user who access to the named database
+  config :password, :validate => :password, :default => nil, :required => true
+
+  # Series name - supports sprintf formatting
+  config :series, :validate => :string, :default => "logstash"
+
+  # Hash of key/value pairs representing data points to send to the named database
+  # Example: `{'column1' => 'value1', 'column2' => 'value2'}`
+  #
+  # Events for the same series will be batched together where possible
+  # Both keys and values support sprintf formatting
+  config :data_points, :validate => :hash, :default => {}, :required => true
+
+  # Allow the override of the `time` column in the event?
+  #
+  # By default any column with a name of `time` will be ignored and the time will
+  # be determined by the value of `@timestamp`.
+  #
+  # Setting this to `true` allows you to explicitly set the `time` column yourself
+  #
+  # Note: **`time` must be an epoch value in either seconds, milliseconds or microseconds**
+  config :allow_time_override, :validate => :boolean, :default => false
+
+  # Set the level of precision of `time`
+  #
+  # only useful when overriding the time value
+  config :time_precision, :validate => ["m", "s", "u"], :default => "s"
+
+  # Allow value coercion
+  #
+  # this will attempt to convert data point values to the appropriate type before posting
+  # otherwise sprintf-filtered numeric values could get sent as strings
+  # format is `{'column_name' => 'datatype'}
+  #
+  # currently supported datatypes are integer and float
+  #
+  config :coerce_values, :validate => :hash, :default => {}
+
+  # This setting controls how many events will be buffered before sending a batch
+  # of events. Note that these are only batched for the same series
+  config :flush_size, :validate => :number, :default => 100
+
+  # The amount of time since last flush before a flush is forced.
+  #
+  # This setting helps ensure slow event rates don't get stuck in Logstash.
+  # For example, if your `flush_size` is 100, and you have received 10 events,
+  # and it has been more than `idle_flush_time` seconds since the last flush,
+  # logstash will flush those 10 events automatically.
+  #
+  # This helps keep both fast and slow log streams moving along in
+  # near-real-time.
+  config :idle_flush_time, :validate => :number, :default => 1
+
+  public
+  def register
+    require "ftw" # gem ftw
+    require 'cgi'
+    @agent = FTW::Agent.new
+    @queue = []
+
+    @query_params = "u=#{@user}&p=#{@password.value}&time_precision=#{@time_precision}"
+    @base_url = "http://#{@host}:#{@port}/db/#{@db}/series"
+    @url = "#{@base_url}?#{@query_params}"
+    
+    buffer_initialize(
+      :max_items => @flush_size,
+      :max_interval => @idle_flush_time,
+      :logger => @logger
+    )
+  end # def register
+  
+  public
+  def receive(event)
+    return unless output?(event)
+
+    # A batch POST for InfluxDB looks like this:
+    # [
+    #   {
+    #     "name": "events",
+    #     "columns": ["state", "email", "type"],
+    #     "points": [
+    #       ["ny", "paul@influxdb.org", "follow"],
+    #       ["ny", "todd@influxdb.org", "open"]
+    #     ]
+    #   },
+    #   {
+    #     "name": "errors",
+    #     "columns": ["class", "file", "user", "severity"],
+    #     "points": [
+    #       ["DivideByZero", "example.py", "someguy@influxdb.org", "fatal"]
+    #     ]
+    #   }
+    # ]
+    event_hash = {}
+    event_hash['name'] = event.sprintf(@series)
+    sprintf_points = Hash[@data_points.map {|k,v| [event.sprintf(k), event.sprintf(v)]}]
+    if sprintf_points.has_key?('time')
+      @logger.error("Cannot override value of time without 'allow_override_time'. Using event timestamp") unless @allow_override_time
+    else
+      sprintf_points['time'] = to_epoch(event.timestamp)
+    end
+    @coerce_values.each do |column, value_type|
+      if sprintf_points.has_key?(column)
+        begin
+          case value_type
+          when "integer"
+            @logger.debug("Converting column #{column} to type #{value_type}: Current value: #{sprintf_points[column]}")
+            sprintf_points[column] = sprintf_points[column].to_i
+          when "float"
+            @logger.debug("Converting column #{column} to type #{value_type}: Current value: #{sprintf_points[column]}")
+            sprintf_points[column] = sprintf_points[column].to_f
+          else
+            @logger.error("Don't know how to convert to #{value_type}")
+          end
+        rescue => e
+          @logger.error("Unhandled exception", :error => e.message)
+        end
+      end
+    end
+    event_hash['columns'] = sprintf_points.keys
+    event_hash['points'] = []
+    event_hash['points'] << sprintf_points.values
+    buffer_receive(event_hash)
+  end # def receive
+
+#  def flush; return; end
+  def flush(events, teardown=false)
+   # Avoid creating a new string for newline every time
+    newline = "\n".freeze
+
+    # seen_series stores a list of series and associated columns
+    # we've seen for each event
+    # so that we can attempt to batch up points for a given series.
+    #
+    # Columns *MUST* be exactly the same
+    seen_series = {}
+    event_collection = []
+
+    events.each do |ev|
+      begin
+        if seen_series.has_key?(ev['name']) and (seen_series[ev['name']] == ev['columns'])
+          @logger.info("Existing series data found. Appending points to that series")
+          event_collection.select {|h| h['points'] << ev['points'][0] if h['name'] == ev['name']}
+        elsif seen_series.has_key?(ev['name']) and (seen_series[ev['name']] != ev['columns'])
+          @logger.warn("Series '#{ev['name']}' has been seen but columns are different or in a different order. Adding to batch but not under existing series")
+          @logger.warn("Existing series columns were: #{seen_series[ev['name']].join(",")} and event columns were: #{ev['columns'].join(",")}")
+          event_collection << ev
+        else
+          seen_series[ev['name']] = ev['columns']
+          event_collection << ev
+        end
+      rescue
+        @logger.info("Error adding event to collection", :exception => e)
+        next
+      end
+    end
+
+    post(event_collection.to_json)
+  end # def receive_bulk
+
+  def post(body)
+    begin
+      @logger.debug("Post body: #{body}")
+      response = @agent.post!(@url, :body => body)
+    rescue EOFError
+      @logger.warn("EOF while writing request or reading response header from InfluxDB",
+                   :host => @host, :port => @port)
+      return # abort this flush
+    end
+
+    # Consume the body for error checking
+    # This will also free up the connection for reuse.
+    body = ""
+    begin
+      response.read_body { |chunk| body += chunk }
+    rescue EOFError
+      @logger.warn("EOF while reading response body from InfluxDB",
+                   :host => @host, :port => @port)
+      return # abort this flush
+    end
+
+    if response.status != 200
+      @logger.error("Error writing to InfluxDB",
+                    :response => response, :response_body => body,
+                    :request_body => @queue.join("\n"))
+      return
+    end
+  end # def post
+
+  private
+  def to_epoch(t)
+    return t.is_a?(Time) ? t.to_i : Time.parse(t).to_i
+  end
+
+  def teardown
+    buffer_flush(:final => true)
+  end # def teardown
+end # class LogStash::Outputs::InfluxDB

data/logstash-output-influxdb.gemspec

@@ -0,0 +1,29 @@
+Gem::Specification.new do |s|
+
+  s.name            = 'logstash-output-influxdb'
+  s.version         = '0.1.0'
+  s.licenses        = ['Apache License (2.0)']
+  s.summary         = "This output lets you output Metrics to InfluxDB"
+  s.description     = "This output lets you output Metrics to InfluxDB"
+  s.authors         = ["Elasticsearch"]
+  s.email           = 'richard.pijnenburg@elasticsearch.com'
+  s.homepage        = "http://logstash.net/"
+  s.require_paths = ["lib"]
+
+  # Files
+  s.files = `git ls-files`.split($\)+::Dir.glob('vendor/*')
+
+  # Tests
+  s.test_files = s.files.grep(%r{^(test|spec|features)/})
+
+  # Special flag to let us know this is actually a logstash plugin
+  s.metadata = { "logstash_plugin" => "true", "group" => "output" }
+
+  # Gem dependencies
+  s.add_runtime_dependency 'logstash', '>= 1.4.0', '< 2.0.0'
+
+  s.add_runtime_dependency 'stud'
+  s.add_runtime_dependency 'ftw', ['~> 0.0.40']
+
+end
+

data/rakelib/publish.rake

@@ -0,0 +1,9 @@
+require "gem_publisher"
+
+desc "Publish gem to RubyGems.org"
+task :publish_gem do |t|
+  gem_file = Dir.glob(File.expand_path('../*.gemspec',File.dirname(__FILE__))).first
+  gem = GemPublisher.publish_if_updated(gem_file, :rubygems)
+  puts "Published #{gem}" if gem
+end
+

data/rakelib/vendor.rake

@@ -0,0 +1,169 @@
+require "net/http"
+require "uri"
+require "digest/sha1"
+
+def vendor(*args)
+  return File.join("vendor", *args)
+end
+
+directory "vendor/" => ["vendor"] do |task, args|
+  mkdir task.name
+end
+
+def fetch(url, sha1, output)
+
+  puts "Downloading #{url}"
+  actual_sha1 = download(url, output)
+
+  if actual_sha1 != sha1
+    fail "SHA1 does not match (expected '#{sha1}' but got '#{actual_sha1}')"
+  end
+end # def fetch
+
+def file_fetch(url, sha1)
+  filename = File.basename( URI(url).path )
+  output = "vendor/#{filename}"
+  task output => [ "vendor/" ] do
+    begin
+      actual_sha1 = file_sha1(output)
+      if actual_sha1 != sha1
+        fetch(url, sha1, output)
+      end
+    rescue Errno::ENOENT
+      fetch(url, sha1, output)
+    end
+  end.invoke
+
+  return output
+end
+
+def file_sha1(path)
+  digest = Digest::SHA1.new
+  fd = File.new(path, "r")
+  while true
+    begin
+      digest << fd.sysread(16384)
+    rescue EOFError
+      break
+    end
+  end
+  return digest.hexdigest
+ensure
+  fd.close if fd
+end
+
+def download(url, output)
+  uri = URI(url)
+  digest = Digest::SHA1.new
+  tmp = "#{output}.tmp"
+  Net::HTTP.start(uri.host, uri.port, :use_ssl => (uri.scheme == "https")) do |http|
+    request = Net::HTTP::Get.new(uri.path)
+    http.request(request) do |response|
+      fail "HTTP fetch failed for #{url}. #{response}" if [200, 301].include?(response.code)
+      size = (response["content-length"].to_i || -1).to_f
+      count = 0
+      File.open(tmp, "w") do |fd|
+        response.read_body do |chunk|
+          fd.write(chunk)
+          digest << chunk
+          if size > 0 && $stdout.tty?
+            count += chunk.bytesize
+            $stdout.write(sprintf("\r%0.2f%%", count/size * 100))
+          end
+        end
+      end
+      $stdout.write("\r      \r") if $stdout.tty?
+    end
+  end
+
+  File.rename(tmp, output)
+
+  return digest.hexdigest
+rescue SocketError => e
+  puts "Failure while downloading #{url}: #{e}"
+  raise
+ensure
+  File.unlink(tmp) if File.exist?(tmp)
+end # def download
+
+def untar(tarball, &block)
+  require "archive/tar/minitar"
+  tgz = Zlib::GzipReader.new(File.open(tarball))
+  # Pull out typesdb
+  tar = Archive::Tar::Minitar::Input.open(tgz)
+  tar.each do |entry|
+    path = block.call(entry)
+    next if path.nil?
+    parent = File.dirname(path)
+    
+    mkdir_p parent unless File.directory?(parent)
+
+    # Skip this file if the output file is the same size
+    if entry.directory?
+      mkdir path unless File.directory?(path)
+    else
+      entry_mode = entry.instance_eval { @mode } & 0777
+      if File.exists?(path)
+        stat = File.stat(path)
+        # TODO(sissel): Submit a patch to archive-tar-minitar upstream to
+        # expose headers in the entry.
+        entry_size = entry.instance_eval { @size }
+        # If file sizes are same, skip writing.
+        next if stat.size == entry_size && (stat.mode & 0777) == entry_mode
+      end
+      puts "Extracting #{entry.full_name} from #{tarball} #{entry_mode.to_s(8)}"
+      File.open(path, "w") do |fd|
+        # eof? check lets us skip empty files. Necessary because the API provided by
+        # Archive::Tar::Minitar::Reader::EntryStream only mostly acts like an
+        # IO object. Something about empty files in this EntryStream causes
+        # IO.copy_stream to throw "can't convert nil into String" on JRuby
+        # TODO(sissel): File a bug about this.
+        while !entry.eof?
+          chunk = entry.read(16384)
+          fd.write(chunk)
+        end
+          #IO.copy_stream(entry, fd)
+      end
+      File.chmod(entry_mode, path)
+    end
+  end
+  tar.close
+  File.unlink(tarball) if File.file?(tarball)
+end # def untar
+
+def ungz(file)
+
+  outpath = file.gsub('.gz', '')
+  tgz = Zlib::GzipReader.new(File.open(file))
+  begin
+    File.open(outpath, "w") do |out|
+      IO::copy_stream(tgz, out)
+    end
+    File.unlink(file)
+  rescue
+    File.unlink(outpath) if File.file?(outpath)
+   raise
+  end
+  tgz.close
+end
+
+desc "Process any vendor files required for this plugin"
+task "vendor" do |task, args|
+
+  @files.each do |file| 
+    download = file_fetch(file['url'], file['sha1'])
+    if download =~ /.tar.gz/
+      prefix = download.gsub('.tar.gz', '').gsub('vendor/', '')
+      untar(download) do |entry|
+        if !file['files'].nil?
+          next unless file['files'].include?(entry.full_name.gsub(prefix, ''))
+          out = entry.full_name.split("/").last
+        end
+        File.join('vendor', out)
+      end
+    elsif download =~ /.gz/
+      ungz(download)
+    end
+  end
+
+end

data/spec/outputs/influxdb_spec.rb

@@ -0,0 +1 @@
+require 'spec_helper'

metadata

@@ -0,0 +1,103 @@
+--- !ruby/object:Gem::Specification
+name: logstash-output-influxdb
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Elasticsearch
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-11-06 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: logstash
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 1.4.0
+    - - <
+      - !ruby/object:Gem::Version
+        version: 2.0.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: 1.4.0
+    - - <
+      - !ruby/object:Gem::Version
+        version: 2.0.0
+- !ruby/object:Gem::Dependency
+  name: stud
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: ftw
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 0.0.40
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 0.0.40
+description: This output lets you output Metrics to InfluxDB
+email: richard.pijnenburg@elasticsearch.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- LICENSE
+- Rakefile
+- lib/logstash/outputs/influxdb.rb
+- logstash-output-influxdb.gemspec
+- rakelib/publish.rake
+- rakelib/vendor.rake
+- spec/outputs/influxdb_spec.rb
+homepage: http://logstash.net/
+licenses:
+- Apache License (2.0)
+metadata:
+  logstash_plugin: 'true'
+  group: output
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.4.1
+signing_key: 
+specification_version: 4
+summary: This output lets you output Metrics to InfluxDB
+test_files:
+- spec/outputs/influxdb_spec.rb