logstash-input-s3 0.1.0

checksums.yaml

@@ -0,0 +1,15 @@
+---
+!binary "U0hBMQ==":
+  metadata.gz: !binary |-
+    NzkxNDE2MTI2NWM5YzVjNGNlNzU5ZjI3OTMxM2U2OWVkZjQ1ZDJmOQ==
+  data.tar.gz: !binary |-
+    YzlhZWI4NGMyZTY1N2E0Njk5M2Y3NjI1YTIwNzYyNWJhM2QzZDI2ZQ==
+SHA512:
+  metadata.gz: !binary |-
+    ZDkzNTVhYmE2YTg4ZWYyZTNiMjQ5M2RhMjRiNmZiNzE0ZDFhZjIxMTIwNDM0
+    NmFhMzMyZmY2MTkwZWM3Y2ZlZDhiYzQ1NThhNzQ2ODgyYThjNzFiNDhkMmE3
+    Zjc5N2U2ODlkYmYwM2RiZmQ4Y2I0OGVmZTc0YTdhN2IzMGI3YmU=
+  data.tar.gz: !binary |-
+    NjI1NmFiOTY1ZWNiNGZiZTQ4YjE4ZjFiNGY3ZjU3MGU1NDgwYTQyODI0YTEx
+    MGNkMjA5MGNmNTdmNTJlNDAyNjlhODc4ZDkzYWFkNWM4MzU3ZTJmZjZlZDdk
+    Y2I5ODIxM2RiYmEwNTA4ZmU5ZGQyN2YxZjMzNjc4NzEyYWM3NmM=

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/inputs/s3.rb

@@ -0,0 +1,278 @@
+# encoding: utf-8
+require "logstash/inputs/base"
+require "logstash/namespace"
+
+require "time"
+require "tmpdir"
+
+# 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.
+class LogStash::Inputs::S3 < LogStash::Inputs::Base
+  config_name "s3"
+  milestone 1
+
+  # TODO(sissel): refactor to use 'line' codec (requires removing both gzip
+  # support and readline usage). Support gzip through a gzip codec! ;)
+  default :codec, "plain"
+
+  # The credentials of the AWS account used to access the bucket.
+  # Credentials can be specified:
+  # - As an ["id","secret"] array
+  # - As a path to a file containing AWS_ACCESS_KEY_ID=... and AWS_SECRET_ACCESS_KEY=...
+  # - In the environment, if not set (using variables AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY)
+  config :credentials, :validate => :array, :default => []
+
+  # The name of the S3 bucket.
+  config :bucket, :validate => :string, :required => true
+
+  # The AWS region for your bucket.
+  config :region, :validate => ["us-east-1", "us-west-1", "us-west-2",
+                                "eu-west-1", "ap-southeast-1", "ap-southeast-2",
+                                "ap-northeast-1", "sa-east-1", "us-gov-west-1"],
+                                :deprecated => "'region' has been deprecated in favor of 'region_endpoint'"
+
+  # The AWS region for your bucket.
+  config :region_endpoint, :validate => ["us-east-1", "us-west-1", "us-west-2",
+                                "eu-west-1", "ap-southeast-1", "ap-southeast-2",
+                                "ap-northeast-1", "sa-east-1", "us-gov-west-1"], :default => "us-east-1"
+
+  # If specified, the prefix the filenames in the bucket must match (not a regexp)
+  config :prefix, :validate => :string, :default => nil
+
+  # Where to write the since database (keeps track of the date
+  # the last handled file was added to S3). The default will write
+  # sincedb files to some path matching "$HOME/.sincedb*"
+  config :sincedb_path, :validate => :string, :default => nil
+
+  # Name of a S3 bucket to backup processed files to.
+  config :backup_to_bucket, :validate => :string, :default => nil
+
+  # Path of a local directory to backup processed files to.
+  config :backup_to_dir, :validate => :string, :default => nil
+
+  # Whether to delete processed files from the original bucket.
+  config :delete, :validate => :boolean, :default => false
+
+  # Interval to wait between to check the file list again after a run is finished.
+  # Value is in seconds.
+  config :interval, :validate => :number, :default => 60
+
+  public
+  def register
+    require "digest/md5"
+    require "aws-sdk"
+
+    @region_endpoint = @region if @region && !@region.empty?
+
+    @logger.info("Registering s3 input", :bucket => @bucket, :region_endpoint => @region_endpoint)
+
+    if @credentials.length == 0
+      @access_key_id = ENV['AWS_ACCESS_KEY_ID']
+      @secret_access_key = ENV['AWS_SECRET_ACCESS_KEY']
+    elsif @credentials.length == 1
+      File.open(@credentials[0]) { |f| f.each do |line|
+        unless (/^\#/.match(line))
+          if(/\s*=\s*/.match(line))
+            param, value = line.split('=', 2)
+            param = param.chomp().strip()
+            value = value.chomp().strip()
+            if param.eql?('AWS_ACCESS_KEY_ID')
+              @access_key_id = value
+            elsif param.eql?('AWS_SECRET_ACCESS_KEY')
+              @secret_access_key = value
+            end
+          end
+        end
+      end
+      }
+    elsif @credentials.length == 2
+      @access_key_id = @credentials[0]
+      @secret_access_key = @credentials[1]
+    else
+      raise ArgumentError.new('Credentials must be of the form "/path/to/file" or ["id", "secret"]')
+    end
+
+    if @access_key_id.nil? or @secret_access_key.nil?
+      raise ArgumentError.new('Missing AWS credentials')
+    end
+
+    if @bucket.nil?
+      raise ArgumentError.new('Missing AWS bucket')
+    end
+
+    if @sincedb_path.nil?
+      if ENV['HOME'].nil?
+        raise ArgumentError.new('No HOME or sincedb_path set')
+      end
+      @sincedb_path = File.join(ENV["HOME"], ".sincedb_" + Digest::MD5.hexdigest("#{@bucket}+#{@prefix}"))
+    end
+
+    s3 = AWS::S3.new(
+      :access_key_id => @access_key_id,
+      :secret_access_key => @secret_access_key,
+      :region => @region_endpoint
+    )
+
+    @s3bucket = s3.buckets[@bucket]
+
+    unless @backup_to_bucket.nil?
+      @backup_bucket = s3.buckets[@backup_to_bucket]
+      unless @backup_bucket.exists?
+        s3.buckets.create(@backup_to_bucket)
+      end
+    end
+
+    unless @backup_to_dir.nil?
+      Dir.mkdir(@backup_to_dir, 0700) unless File.exists?(@backup_to_dir)
+    end
+
+  end # def register
+
+  public
+  def run(queue)
+    loop do
+      process_new(queue)
+      sleep(@interval)
+    end
+    finished
+  end # def run
+
+  private
+  def process_new(queue, since=nil)
+
+    if since.nil?
+        since = sincedb_read()
+    end
+
+    objects = list_new(since)
+    objects.each do |k|
+      @logger.debug("S3 input processing", :bucket => @bucket, :key => k)
+      lastmod = @s3bucket.objects[k].last_modified
+      process_log(queue, k)
+      sincedb_write(lastmod)
+    end
+
+  end # def process_new
+
+  private
+  def list_new(since=nil)
+
+    if since.nil?
+      since = Time.new(0)
+    end
+
+    objects = {}
+    @s3bucket.objects.with_prefix(@prefix).each do |log|
+      if log.last_modified > since
+        objects[log.key] = log.last_modified
+      end
+    end
+
+    return sorted_objects = objects.keys.sort {|a,b| objects[a] <=> objects[b]}
+
+  end # def list_new
+
+  private
+  def process_log(queue, key)
+
+    object = @s3bucket.objects[key]
+    tmp = Dir.mktmpdir("logstash-")
+    begin
+      filename = File.join(tmp, File.basename(key))
+      File.open(filename, 'wb') do |s3file|
+        object.read do |chunk|
+          s3file.write(chunk)
+        end
+      end
+      process_local_log(queue, filename)
+      unless @backup_to_bucket.nil?
+        backup_object = @backup_bucket.objects[key]
+        backup_object.write(Pathname.new(filename))
+      end
+      unless @backup_to_dir.nil?
+        FileUtils.cp(filename, @backup_to_dir)
+      end
+      if @delete
+        object.delete()
+      end
+    end
+    FileUtils.remove_entry_secure(tmp, force=true)
+
+  end # def process_log
+
+  private
+  def process_local_log(queue, filename)
+
+    metadata = {
+      :version => nil,
+      :format => nil,
+    }
+    File.open(filename) do |file|
+      if filename.end_with?('.gz')
+        gz = Zlib::GzipReader.new(file)
+        gz.each_line do |line|
+          metadata = process_line(queue, metadata, line)
+        end
+      else
+        file.each do |line|
+          metadata = process_line(queue, metadata, line)
+        end
+      end
+    end
+
+  end # def process_local_log
+
+  private
+  def process_line(queue, metadata, line)
+
+    if /#Version: .+/.match(line)
+      junk, version = line.strip().split(/#Version: (.+)/)
+      unless version.nil?
+        metadata[:version] = version
+      end
+    elsif /#Fields: .+/.match(line)
+      junk, format = line.strip().split(/#Fields: (.+)/)
+      unless format.nil?
+        metadata[:format] = format
+      end
+    else
+      @codec.decode(line) do |event|
+        decorate(event)
+        unless metadata[:version].nil?
+          event["cloudfront_version"] = metadata[:version]
+        end
+        unless metadata[:format].nil?
+          event["cloudfront_fields"] = metadata[:format]
+        end
+        queue << event
+      end
+    end
+    return metadata
+
+  end # def process_line
+
+  private
+  def sincedb_read()
+
+    if File.exists?(@sincedb_path)
+      since = Time.parse(File.read(@sincedb_path).chomp.strip)
+    else
+      since = Time.new(0)
+    end
+    return since
+
+  end # def sincedb_read
+
+  private
+  def sincedb_write(since=nil)
+
+    if since.nil?
+      since = Time.now()
+    end
+    File.open(@sincedb_path, 'w') { |file| file.write(since.to_s) }
+
+  end # def sincedb_write
+
+end # class LogStash::Inputs::S3

data/logstash-input-s3.gemspec

@@ -0,0 +1,29 @@
+Gem::Specification.new do |s|
+
+  s.name            = 'logstash-input-s3'
+  s.version         = '0.1.0'
+  s.licenses        = ['Apache License (2.0)']
+  s.summary         = "Stream events from files from a S3 bucket."
+  s.description     = "Stream events from files from a S3 bucket."
+  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" => "input" }
+
+  # Gem dependencies
+  s.add_runtime_dependency 'logstash', '>= 1.4.0', '< 2.0.0'
+
+  s.add_runtime_dependency 'logstash-codec-plain'
+  s.add_runtime_dependency 'aws-sdk'
+
+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/inputs/s3_spec.rb

@@ -0,0 +1,5 @@
+require 'spec_helper'
+require 'logstash/inputs/s3'
+
+describe LogStash::Inputs::S3 do
+end

metadata

@@ -0,0 +1,103 @@
+--- !ruby/object:Gem::Specification
+name: logstash-input-s3
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Elasticsearch
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2014-11-05 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: logstash-codec-plain
+  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: aws-sdk
+  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'
+description: Stream events from files from a S3 bucket.
+email: richard.pijnenburg@elasticsearch.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- LICENSE
+- Rakefile
+- lib/logstash/inputs/s3.rb
+- logstash-input-s3.gemspec
+- rakelib/publish.rake
+- rakelib/vendor.rake
+- spec/inputs/s3_spec.rb
+homepage: http://logstash.net/
+licenses:
+- Apache License (2.0)
+metadata:
+  logstash_plugin: 'true'
+  group: input
+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: Stream events from files from a S3 bucket.
+test_files:
+- spec/inputs/s3_spec.rb