logstash-input-s3 3.5.0 → 3.8.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a6d9ab89a4d7925dbaaa02b021b1bbe803426a5c2e5285c1239d72950563fc27
-  data.tar.gz: 40aafdb8002e940fcc08f72d119299567dda77210dedcaf436df9a273858ecf1
+  metadata.gz: dd11c4e3d6df5c28143a4a67de1dcaddb7f3d9bd7a0921c278d15482634bc676
+  data.tar.gz: c9687036e353d7a8a71e4d4ec85e497a110ea740b22b6b98173350bcafe8f5b4
 SHA512:
-  metadata.gz: 12730fa07325e2549ac32c8b7a629464c3a1d789992c5e07bd3f1bf43ad11b353886882678d00c27a14a7e0e675eaaa4002187f5141d078308de8e3e480d67d3
-  data.tar.gz: 24c61bb4d995ef2615cf9ea769ad36d88965eb758e8a9d205394a887ca6f1cac47211e76c42c81e8499f1ab818e072ee50e52431de045e1018402af947f61df3
+  metadata.gz: 3e552d9501b11be011db52415f446a00fc7bfe2ecb601a50f3fa5938de1931fadaded204e46626fd530b6779802627e92f7fe56cda6db5b7f88cd5efadfec2a7
+  data.tar.gz: ed566e40ac8bb1acb29751b222453fa01a315b40a0d3778d23cc024db3d303fd8baf09d287b91f1492d701c007d972032003eb04f2a45c11ab0fe07dff21c77b

data/CHANGELOG.md

@@ -1,3 +1,22 @@
+## 3.8.1
+ - Feat: cast true/false values for additional_settings [#232](https://github.com/logstash-plugins/logstash-input-s3/pull/232)
+
+## 3.8.0
+ - Add ECS v8 support.
+
+## 3.7.0
+ - Add ECS support. [#228](https://github.com/logstash-plugins/logstash-input-s3/pull/228)
+ - Fix missing file in cutoff time change. [#224](https://github.com/logstash-plugins/logstash-input-s3/pull/224)
+
+## 3.6.0
+ - Fixed unprocessed file with the same `last_modified` in ingestion. [#220](https://github.com/logstash-plugins/logstash-input-s3/pull/220)
+
+## 3.5.2
+ - [DOC]Added note that only AWS S3 is supported. No other S3 compatible storage solutions are supported. [#208](https://github.com/logstash-plugins/logstash-input-s3/issues/208)
+
+## 3.5.1
+ - [DOC]Added example for `exclude_pattern` and reordered option descriptions [#204](https://github.com/logstash-plugins/logstash-input-s3/issues/204)
+ 
 ## 3.5.0
  - Added support for including objects restored from Glacier or Glacier Deep [#199](https://github.com/logstash-plugins/logstash-input-s3/issues/199)
  - Added `gzip_pattern` option, enabling more flexible determination of whether a file is gzipped [#165](https://github.com/logstash-plugins/logstash-input-s3/issues/165)

data/README.md

@@ -1,6 +1,6 @@
 # Logstash Plugin
 
-[![Travis Build Status](https://travis-ci.org/logstash-plugins/logstash-input-s3.svg)](https://travis-ci.org/logstash-plugins/logstash-input-s3)
+[![Travis Build Status](https://travis-ci.com/logstash-plugins/logstash-input-s3.svg)](https://travis-ci.com/logstash-plugins/logstash-input-s3)
 
 This is a plugin for [Logstash](https://github.com/elastic/logstash).
 

data/docs/index.asciidoc

@@ -23,11 +23,29 @@ include::{include_path}/plugin_header.asciidoc[]
 
 Stream events from files from a S3 bucket.
 
+IMPORTANT: The S3 input plugin only supports AWS S3. 
+Other S3 compatible storage solutions are not supported.
+
 Each line from each file generates an event.
 Files ending in `.gz` are handled as gzip'ed files.
 
 Files that are archived to AWS Glacier will be skipped.
 
+[id="plugins-{type}s-{plugin}-ecs_metadata"]
+==== Event Metadata and the Elastic Common Schema (ECS)
+This plugin adds cloudfront metadata to event.
+When ECS compatibility is disabled, the value is stored in the root level.
+When ECS is enabled, the value is stored in the `@metadata` where it can be used by other plugins in your pipeline.
+
+Here’s how ECS compatibility mode affects output.
+[cols="<l,<l,e,<e"]
+|=======================================================================
+| ECS disabled | ECS v1 | Availability | Description
+
+| cloudfront_fields | [@metadata][s3][cloudfront][fields] | available when the file is a CloudFront log | column names of log
+| cloudfront_version | [@metadata][s3][cloudfront][version] | available when the file is a CloudFront log | version of log
+|=======================================================================
+
 [id="plugins-{type}s-{plugin}-options"]
 ==== S3 Input Configuration Options
 
@@ -44,6 +62,7 @@ This plugin supports the following configuration options plus the <<plugins-{typ
 | <<plugins-{type}s-{plugin}-backup_to_dir>> |<<string,string>>|No
 | <<plugins-{type}s-{plugin}-bucket>> |<<string,string>>|Yes
 | <<plugins-{type}s-{plugin}-delete>> |<<boolean,boolean>>|No
+| <<plugins-{type}s-{plugin}-ecs_compatibility>> |<<string,string>>|No
 | <<plugins-{type}s-{plugin}-endpoint>> |<<string,string>>|No
 | <<plugins-{type}s-{plugin}-exclude_pattern>> |<<string,string>>|No
 | <<plugins-{type}s-{plugin}-gzip_pattern>> |<<string,string>>|No
@@ -80,6 +99,29 @@ This plugin uses the AWS SDK and supports several ways to get credentials, which
 4. Environment variables `AMAZON_ACCESS_KEY_ID` and `AMAZON_SECRET_ACCESS_KEY`
 5. IAM Instance Profile (available when running inside EC2)
 
+
+[id="plugins-{type}s-{plugin}-additional_settings"]
+===== `additional_settings`
+
+  * Value type is <<hash,hash>>
+  * Default value is `{}`
+
+Key-value pairs of settings and corresponding values used to parametrize
+the connection to s3. See full list in https://docs.aws.amazon.com/sdkforruby/api/Aws/S3/Client.html[the AWS SDK documentation]. Example:
+
+[source,ruby]
+    input {
+      s3 {
+        access_key_id => "1234"
+        secret_access_key => "secret"
+        bucket => "logstash-test"
+        additional_settings => {
+          force_path_style => true
+          follow_redirects => false
+        }
+      }
+    }
+
 [id="plugins-{type}s-{plugin}-aws_credentials_file"]
 ===== `aws_credentials_file` 
 
@@ -141,6 +183,18 @@ The name of the S3 bucket.
 
 Whether to delete processed files from the original bucket.
 
+[id="plugins-{type}s-{plugin}-ecs_compatibility"]
+===== `ecs_compatibility`
+
+  * Value type is <<string,string>>
+  * Supported values are:
+  ** `disabled`: does not use ECS-compatible field names
+  ** `v1`,`v8`: uses metadata fields that are compatible with Elastic Common Schema
+
+Controls this plugin's compatibility with the
+{ecs-ref}[Elastic Common Schema (ECS)].
+See <<plugins-{type}s-{plugin}-ecs_metadata>> for detailed information.
+
 [id="plugins-{type}s-{plugin}-endpoint"]
 ===== `endpoint`
 
@@ -157,7 +211,20 @@ guaranteed to work correctly with the AWS SDK.
   * Value type is <<string,string>>
   * Default value is `nil`
 
-Ruby style regexp of keys to exclude from the bucket
+Ruby style regexp of keys to exclude from the bucket.
+
+Note that files matching the pattern are skipped _after_ they have been listed. 
+Consider using <<plugins-{type}s-{plugin}-prefix>> instead where possible.
+
+Example:
+
+[source,ruby]
+-----
+"exclude_pattern" => "\/2020\/04\/"
+-----
+
+This pattern excludes all logs containing "/2020/04/" in the path.
+
 
 [id="plugins-{type}s-{plugin}-gzip_pattern"]
 ===== `gzip_pattern`
@@ -167,28 +234,6 @@ Ruby style regexp of keys to exclude from the bucket
 
 Regular expression used to determine whether an input file is in gzip format.
 
-[id="plugins-{type}s-{plugin}-additional_settings"]
-===== `additional_settings`
-
-  * Value type is <<hash,hash>>
-  * Default value is `{}`
-
-Key-value pairs of settings and corresponding values used to parametrize
-the connection to s3. See full list in https://docs.aws.amazon.com/sdkforruby/api/Aws/S3/Client.html[the AWS SDK documentation]. Example:
-
-[source,ruby]
-    input {
-      s3 {
-        "access_key_id" => "1234"
-        "secret_access_key" => "secret"
-        "bucket" => "logstash-test"
-        "additional_settings" => {
-          "force_path_style" => true
-          "follow_redirects" => false
-        }
-      }
-    }
-
 [id="plugins-{type}s-{plugin}-include_object_properties"]
 ===== `include_object_properties`
 

data/lib/logstash/inputs/s3.rb

@@ -9,6 +9,7 @@ require "stud/interval"
 require "stud/temporary"
 require "aws-sdk"
 require "logstash/inputs/s3/patch"
+require "logstash/plugin_mixins/ecs_compatibility_support"
 
 require 'java'
 
@@ -27,6 +28,7 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
   java_import java.util.zip.ZipException
 
   include LogStash::PluginMixins::AwsConfig::V2
+  include LogStash::PluginMixins::ECSCompatibilitySupport(:disabled, :v1, :v8 => :v1)
 
   config_name "s3"
 
@@ -86,6 +88,14 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
   # default to an expression that matches *.gz and *.gzip file extensions
   config :gzip_pattern, :validate => :string, :default => "\.gz(ip)?$"
 
+  CUTOFF_SECOND = 3
+
+  def initialize(*params)
+    super
+    @cloudfront_fields_key = ecs_select[disabled: 'cloudfront_fields', v1: '[@metadata][s3][cloudfront][fields]']
+    @cloudfront_version_key = ecs_select[disabled: 'cloudfront_version', v1: '[@metadata][s3][cloudfront][version]']
+  end
+
   def register
     require "fileutils"
     require "digest/md5"
@@ -126,8 +136,9 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
   end # def run
 
   def list_new_files
-    objects = {}
+    objects = []
     found = false
+    current_time = Time.now
     begin
       @s3bucket.objects(:prefix => @prefix).each do |log|
         found = true
@@ -138,10 +149,12 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
           @logger.debug('Object Zero Length', :key => log.key)
         elsif !sincedb.newer?(log.last_modified)
           @logger.debug('Object Not Modified', :key => log.key)
+        elsif log.last_modified > (current_time - CUTOFF_SECOND).utc # file modified within last two seconds will be processed in next cycle
+          @logger.debug('Object Modified After Cutoff Time', :key => log.key)
         elsif (log.storage_class == 'GLACIER' || log.storage_class == 'DEEP_ARCHIVE') && !file_restored?(log.object)
           @logger.debug('Object Archived to Glacier', :key => log.key)
         else
-          objects[log.key] = log.last_modified
+          objects << log
           @logger.debug("Added to objects[]", :key => log.key, :length => objects.length)
         end
       end
@@ -149,7 +162,7 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
     rescue Aws::Errors::ServiceError => e
       @logger.error("Unable to list objects in bucket", :exception => e.class, :message => e.message, :backtrace => e.backtrace, :prefix => prefix)
     end
-    objects.keys.sort {|a,b| objects[a] <=> objects[b]}
+    objects.sort_by { |log| log.last_modified }
   end # def fetch_new_files
 
   def backup_to_bucket(object)
@@ -171,11 +184,11 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
   def process_files(queue)
     objects = list_new_files
 
-    objects.each do |key|
+    objects.each do |log|
       if stop?
         break
       else
-        process_log(queue, key)
+        process_log(queue, log)
       end
     end
   end # def process_files
@@ -223,9 +236,6 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
         else
           decorate(event)
 
-          event.set("cloudfront_version", metadata[:cloudfront_version]) unless metadata[:cloudfront_version].nil?
-          event.set("cloudfront_fields", metadata[:cloudfront_fields]) unless metadata[:cloudfront_fields].nil?
-
           if @include_object_properties
             event.set("[@metadata][s3]", object.data.to_h)
           else
@@ -233,6 +243,8 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
           end
 
           event.set("[@metadata][s3][key]", object.key)
+          event.set(@cloudfront_version_key, metadata[:cloudfront_version]) unless metadata[:cloudfront_version].nil?
+          event.set(@cloudfront_fields_key, metadata[:cloudfront_fields]) unless metadata[:cloudfront_fields].nil?
 
           queue << event
         end
@@ -341,14 +353,22 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
   end
 
   def symbolized_settings
-    @symbolized_settings ||= symbolize(@additional_settings)
+    @symbolized_settings ||= symbolize_keys_and_cast_true_false(@additional_settings)
   end
 
-  def symbolize(hash)
-    return hash unless hash.is_a?(Hash)
-    symbolized = {}
-    hash.each { |key, value| symbolized[key.to_sym] = symbolize(value) }
-    symbolized
+  def symbolize_keys_and_cast_true_false(hash)
+    case hash
+    when Hash
+      symbolized = {}
+      hash.each { |key, value| symbolized[key.to_sym] = symbolize_keys_and_cast_true_false(value) }
+      symbolized
+    when 'true'
+      true
+    when 'false'
+      false
+    else
+      hash
+    end
   end
 
   def ignore_filename?(filename)
@@ -367,19 +387,22 @@ class LogStash::Inputs::S3 < LogStash::Inputs::Base
     end
   end
 
-  def process_log(queue, key)
-    @logger.debug("Processing", :bucket => @bucket, :key => key)
-    object = @s3bucket.object(key)
+  def process_log(queue, log)
+    @logger.debug("Processing", :bucket => @bucket, :key => log.key)
+    object = @s3bucket.object(log.key)
 
-    filename = File.join(temporary_directory, File.basename(key))
+    filename = File.join(temporary_directory, File.basename(log.key))
     if download_remote_file(object, filename)
       if process_local_log(queue, filename, object)
-        lastmod = object.last_modified
-        backup_to_bucket(object)
-        backup_to_dir(filename)
-        delete_file_from_bucket(object)
-        FileUtils.remove_entry_secure(filename, true)
-        sincedb.write(lastmod)
+        if object.last_modified == log.last_modified
+          backup_to_bucket(object)
+          backup_to_dir(filename)
+          delete_file_from_bucket(object)
+          FileUtils.remove_entry_secure(filename, true)
+          sincedb.write(log.last_modified)
+        else
+          @logger.info("#{log.key} is updated at #{object.last_modified} and will process in the next cycle")
+        end
       end
     else
       FileUtils.remove_entry_secure(filename, true)

data/logstash-input-s3.gemspec

@@ -1,7 +1,7 @@
 Gem::Specification.new do |s|
 
   s.name            = 'logstash-input-s3'
-  s.version         = '3.5.0'
+  s.version         = '3.8.1'
   s.licenses        = ['Apache-2.0']
   s.summary         = "Streams events from files in a S3 bucket"
   s.description     = "This gem is a Logstash plugin required to be installed on top of the Logstash core pipeline using $LS_HOME/bin/logstash-plugin install gemname. This gem is not a stand-alone program"
@@ -27,4 +27,5 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'logstash-devutils'
   s.add_development_dependency "logstash-codec-json"
   s.add_development_dependency "logstash-codec-multiline"
+  s.add_runtime_dependency 'logstash-mixin-ecs_compatibility_support', '~>1.2'
 end

data/spec/inputs/s3_spec.rb

@@ -9,6 +9,7 @@ require_relative "../support/helpers"
 require "stud/temporary"
 require "aws-sdk"
 require "fileutils"
+require 'logstash/plugin_mixins/ecs_compatibility_support/spec_helper'
 
 describe LogStash::Inputs::S3 do
   let(:temporary_directory) { Stud::Temporary.pathname }
@@ -24,6 +25,7 @@ describe LogStash::Inputs::S3 do
       "sincedb_path" => File.join(sincedb_path, ".sincedb")
     }
   }
+  let(:cutoff) { LogStash::Inputs::S3::CUTOFF_SECOND }
 
 
   before do
@@ -33,13 +35,16 @@ describe LogStash::Inputs::S3 do
   end
 
   context "when interrupting the plugin" do
-    let(:config) { super.merge({ "interval" => 5 }) }
+    let(:config) { super().merge({ "interval" => 5 }) }
+    let(:s3_obj) { double(:key => "awesome-key", :last_modified => Time.now.round, :content_length => 10, :storage_class => 'STANDARD', :object => double(:data => double(:restore => nil)) ) }
 
     before do
-      expect_any_instance_of(LogStash::Inputs::S3).to receive(:list_new_files).and_return(TestInfiniteS3Object.new)
+      expect_any_instance_of(LogStash::Inputs::S3).to receive(:list_new_files).and_return(TestInfiniteS3Object.new(s3_obj))
     end
 
-    it_behaves_like "an interruptible input plugin"
+    it_behaves_like "an interruptible input plugin" do
+      let(:allowed_lag) { 16 } if LOGSTASH_VERSION.split('.').first.to_i <= 6
+    end
   end
 
   describe "#register" do
@@ -79,10 +84,10 @@ describe LogStash::Inputs::S3 do
     end
 
     describe "additional_settings" do
-      context 'when force_path_style is set' do
+      context "supported settings" do
         let(:settings) {
           {
-            "additional_settings" => { "force_path_style" => true },
+            "additional_settings" => { "force_path_style" => 'true', "ssl_verify_peer" => 'false', "profile" => 'logstash' },
             "bucket" => "logstash-test",
           }
         }
@@ -90,7 +95,7 @@ describe LogStash::Inputs::S3 do
         it 'should instantiate AWS::S3 clients with force_path_style set' do
           expect(Aws::S3::Resource).to receive(:new).with({
             :region => subject.region,
-            :force_path_style => true
+            :force_path_style => true, :ssl_verify_peer => false, :profile => 'logstash'
           }).and_call_original
 
           subject.send(:get_s3object)
@@ -115,11 +120,12 @@ describe LogStash::Inputs::S3 do
   describe "#list_new_files" do
     before { allow_any_instance_of(Aws::S3::Bucket).to receive(:objects) { objects_list } }
 
-    let!(:present_object) {double(:key => 'this-should-be-present', :last_modified => Time.now, :content_length => 10, :storage_class => 'STANDARD', :object => double(:data => double(:restore => nil)) ) }
-    let!(:archived_object) {double(:key => 'this-should-be-archived', :last_modified => Time.now, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => nil)) ) }
-    let!(:deep_archived_object) {double(:key => 'this-should-be-archived', :last_modified => Time.now, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => nil)) ) }
-    let!(:restored_object) {double(:key => 'this-should-be-restored-from-archive', :last_modified => Time.now, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => 'ongoing-request="false", expiry-date="Thu, 01 Jan 2099 00:00:00 GMT"')) ) }
-    let!(:deep_restored_object) {double(:key => 'this-should-be-restored-from-deep-archive', :last_modified => Time.now, :content_length => 10, :storage_class => 'DEEP_ARCHIVE', :object => double(:data => double(:restore => 'ongoing-request="false", expiry-date="Thu, 01 Jan 2099 00:00:00 GMT"')) ) }
+    let!(:present_object_after_cutoff) {double(:key => 'this-should-not-be-present', :last_modified => Time.now, :content_length => 10, :storage_class => 'STANDARD', :object => double(:data => double(:restore => nil)) ) }
+    let!(:present_object) {double(:key => 'this-should-be-present', :last_modified => Time.now - cutoff, :content_length => 10, :storage_class => 'STANDARD', :object => double(:data => double(:restore => nil)) ) }
+    let!(:archived_object) {double(:key => 'this-should-be-archived', :last_modified => Time.now  - cutoff, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => nil)) ) }
+    let!(:deep_archived_object) {double(:key => 'this-should-be-archived', :last_modified => Time.now - cutoff, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => nil)) ) }
+    let!(:restored_object) {double(:key => 'this-should-be-restored-from-archive', :last_modified => Time.now  - cutoff, :content_length => 10, :storage_class => 'GLACIER', :object => double(:data => double(:restore => 'ongoing-request="false", expiry-date="Thu, 01 Jan 2099 00:00:00 GMT"')) ) }
+    let!(:deep_restored_object) {double(:key => 'this-should-be-restored-from-deep-archive', :last_modified => Time.now  - cutoff, :content_length => 10, :storage_class => 'DEEP_ARCHIVE', :object => double(:data => double(:restore => 'ongoing-request="false", expiry-date="Thu, 01 Jan 2099 00:00:00 GMT"')) ) }
     let(:objects_list) {
       [
         double(:key => 'exclude-this-file-1', :last_modified => Time.now - 2 * day, :content_length => 100, :storage_class => 'STANDARD'),
@@ -127,7 +133,8 @@ describe LogStash::Inputs::S3 do
         archived_object,
         restored_object,
         deep_restored_object,
-        present_object
+        present_object,
+        present_object_after_cutoff
       ]
     }
 
@@ -135,7 +142,7 @@ describe LogStash::Inputs::S3 do
       plugin = LogStash::Inputs::S3.new(config.merge({ "exclude_pattern" => "^exclude" }))
       plugin.register
 
-      files = plugin.list_new_files
+      files = plugin.list_new_files.map { |item| item.key }
       expect(files).to include(present_object.key)
       expect(files).to include(restored_object.key)
       expect(files).to include(deep_restored_object.key)
@@ -143,6 +150,7 @@ describe LogStash::Inputs::S3 do
       expect(files).to_not include('exclude/logstash')    # matches exclude pattern
       expect(files).to_not include(archived_object.key)   # archived
       expect(files).to_not include(deep_archived_object.key)   # archived
+      expect(files).to_not include(present_object_after_cutoff.key)   # after cutoff
       expect(files.size).to eq(3)
     end
 
@@ -150,7 +158,7 @@ describe LogStash::Inputs::S3 do
       plugin = LogStash::Inputs::S3.new(config)
       plugin.register
 
-      files = plugin.list_new_files
+      files = plugin.list_new_files.map { |item| item.key }
       expect(files).to include(present_object.key)
       expect(files).to include(restored_object.key)
       expect(files).to include(deep_restored_object.key)
@@ -158,6 +166,7 @@ describe LogStash::Inputs::S3 do
       expect(files).to include('exclude/logstash')      # no exclude pattern given
       expect(files).to_not include(archived_object.key) # archived
       expect(files).to_not include(deep_archived_object.key)   # archived
+      expect(files).to_not include(present_object_after_cutoff.key)   # after cutoff
       expect(files.size).to eq(5)
     end
 
@@ -204,7 +213,7 @@ describe LogStash::Inputs::S3 do
                                                          'backup_to_bucket' => config['bucket']}))
         plugin.register
 
-        files = plugin.list_new_files
+        files = plugin.list_new_files.map { |item| item.key }
         expect(files).to include(present_object.key)
         expect(files).to_not include('mybackup-log-1') # matches backup prefix
         expect(files.size).to eq(1)
@@ -218,7 +227,7 @@ describe LogStash::Inputs::S3 do
       allow_any_instance_of(LogStash::Inputs::S3::SinceDB::File).to receive(:read).and_return(Time.now - day)
       plugin.register
 
-      files = plugin.list_new_files
+      files = plugin.list_new_files.map { |item| item.key }
       expect(files).to include(present_object.key)
       expect(files).to include(restored_object.key)
       expect(files).to include(deep_restored_object.key)
@@ -226,6 +235,7 @@ describe LogStash::Inputs::S3 do
       expect(files).to_not include('exclude/logstash')    # too old
       expect(files).to_not include(archived_object.key)   # archived
       expect(files).to_not include(deep_archived_object.key) # archived
+      expect(files).to_not include(present_object_after_cutoff.key)   # after cutoff
       expect(files.size).to eq(3)
     end
 
@@ -241,13 +251,14 @@ describe LogStash::Inputs::S3 do
 
         plugin = LogStash::Inputs::S3.new(config.merge({ 'prefix' => prefix }))
         plugin.register
-        expect(plugin.list_new_files).to eq([present_object.key])
+        expect(plugin.list_new_files.map { |item| item.key }).to eq([present_object.key])
     end
 
     it 'should sort return object sorted by last_modification date with older first' do
       objects = [
         double(:key => 'YESTERDAY', :last_modified => Time.now - day, :content_length => 5, :storage_class => 'STANDARD'),
         double(:key => 'TODAY', :last_modified => Time.now, :content_length => 5, :storage_class => 'STANDARD'),
+        double(:key => 'TODAY_BEFORE_CUTOFF', :last_modified => Time.now - cutoff, :content_length => 5, :storage_class => 'STANDARD'),
         double(:key => 'TWO_DAYS_AGO', :last_modified => Time.now - 2 * day, :content_length => 5, :storage_class => 'STANDARD')
       ]
 
@@ -256,7 +267,7 @@ describe LogStash::Inputs::S3 do
 
       plugin = LogStash::Inputs::S3.new(config)
       plugin.register
-      expect(plugin.list_new_files).to eq(['TWO_DAYS_AGO', 'YESTERDAY', 'TODAY'])
+      expect(plugin.list_new_files.map { |item| item.key }).to eq(['TWO_DAYS_AGO', 'YESTERDAY', 'TODAY_BEFORE_CUTOFF'])
     end
 
     describe "when doing backup on the s3" do
@@ -451,7 +462,7 @@ describe LogStash::Inputs::S3 do
     end
 
     context 'compressed with gzip extension and using custom gzip_pattern option' do
-      let(:config) { super.merge({ "gzip_pattern" => "gee.zip$" }) }
+      let(:config) { super().merge({ "gzip_pattern" => "gee.zip$" }) }
       let(:log) { double(:key => 'log.gee.zip', :last_modified => Time.now - 2 * day, :content_length => 5, :storage_class => 'STANDARD') }
       let(:log_file) { File.join(File.dirname(__FILE__), '..', 'fixtures', 'compressed.log.gee.zip') }
        include_examples "generated events"
@@ -486,12 +497,20 @@ describe LogStash::Inputs::S3 do
     context 'cloudfront' do
       let(:log_file) { File.join(File.dirname(__FILE__), '..', 'fixtures', 'cloudfront.log') }
 
-      it 'should extract metadata from cloudfront log' do
-        events = fetch_events(config)
+      describe "metadata", :ecs_compatibility_support, :aggregate_failures do
+        ecs_compatibility_matrix(:disabled, :v1) do |ecs_select|
+          before(:each) do
+            allow_any_instance_of(described_class).to receive(:ecs_compatibility).and_return(ecs_compatibility)
+          end
 
-        events.each do |event|
-          expect(event.get('cloudfront_fields')).to eq('date time x-edge-location c-ip x-event sc-bytes x-cf-status x-cf-client-id cs-uri-stem cs-uri-query c-referrer x-page-url​  c-user-agent x-sname x-sname-query x-file-ext x-sid')
-          expect(event.get('cloudfront_version')).to eq('1.0')
+          it 'should extract metadata from cloudfront log' do
+            events = fetch_events(config)
+
+            events.each do |event|
+              expect(event.get ecs_select[disabled: "cloudfront_fields", v1: "[@metadata][s3][cloudfront][fields]"] ).to eq('date time x-edge-location c-ip x-event sc-bytes x-cf-status x-cf-client-id cs-uri-stem cs-uri-query c-referrer x-page-url​  c-user-agent x-sname x-sname-query x-file-ext x-sid')
+              expect(event.get ecs_select[disabled: "cloudfront_version", v1: "[@metadata][s3][cloudfront][version]"] ).to eq('1.0')
+            end
+          end
         end
       end
 
@@ -499,7 +518,7 @@ describe LogStash::Inputs::S3 do
     end
 
     context 'when include_object_properties is set to true' do
-      let(:config) { super.merge({ "include_object_properties" => true }) }
+      let(:config) { super().merge({ "include_object_properties" => true }) }
       let(:log_file) { File.join(File.dirname(__FILE__), '..', 'fixtures', 'uncompressed.log') }
 
       it 'should extract object properties onto [@metadata][s3]' do
@@ -513,7 +532,7 @@ describe LogStash::Inputs::S3 do
     end
 
     context 'when include_object_properties is set to false' do
-      let(:config) { super.merge({ "include_object_properties" => false }) }
+      let(:config) { super().merge({ "include_object_properties" => false }) }
       let(:log_file) { File.join(File.dirname(__FILE__), '..', 'fixtures', 'uncompressed.log') }
 
       it 'should NOT extract object properties onto [@metadata][s3]' do
@@ -525,6 +544,67 @@ describe LogStash::Inputs::S3 do
 
       include_examples "generated events"
     end
+  end
+
+  describe "data loss" do
+    let(:s3_plugin) { LogStash::Inputs::S3.new(config) }
+    let(:queue) { [] }
+
+    before do
+      s3_plugin.register
+    end
+
+    context 'events come after cutoff time' do
+      it 'should be processed in next cycle' do
+        s3_objects = [
+          double(:key => 'TWO_DAYS_AGO', :last_modified => Time.now.round - 2 * day, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'YESTERDAY', :last_modified => Time.now.round - day, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'TODAY_BEFORE_CUTOFF', :last_modified => Time.now.round - cutoff, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'TODAY', :last_modified => Time.now.round, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'TODAY', :last_modified => Time.now.round, :content_length => 5, :storage_class => 'STANDARD')
+        ]
+        size = s3_objects.length
+
+        allow_any_instance_of(Aws::S3::Bucket).to receive(:objects) { s3_objects }
+        allow_any_instance_of(Aws::S3::Bucket).to receive(:object).and_return(*s3_objects)
+        expect(s3_plugin).to receive(:process_log).at_least(size).and_call_original
+        expect(s3_plugin).to receive(:stop?).and_return(false).at_least(size)
+        expect(s3_plugin).to receive(:download_remote_file).and_return(true).at_least(size)
+        expect(s3_plugin).to receive(:process_local_log).and_return(true).at_least(size)
+
+        # first iteration
+        s3_plugin.process_files(queue)
+
+        # second iteration
+        sleep(cutoff + 1)
+        s3_plugin.process_files(queue)
+      end
+    end
+
+    context 's3 object updated after getting summary' do
+      it 'should not update sincedb' do
+        s3_summary = [
+          double(:key => 'YESTERDAY', :last_modified => Time.now.round - day, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'TODAY', :last_modified => Time.now.round - (cutoff * 10), :content_length => 5, :storage_class => 'STANDARD')
+        ]
 
+        s3_objects = [
+          double(:key => 'YESTERDAY', :last_modified => Time.now.round - day, :content_length => 5, :storage_class => 'STANDARD'),
+          double(:key => 'TODAY_UPDATED', :last_modified => Time.now.round, :content_length => 5, :storage_class => 'STANDARD')
+        ]
+
+        size = s3_objects.length
+
+        allow_any_instance_of(Aws::S3::Bucket).to receive(:objects) { s3_summary }
+        allow_any_instance_of(Aws::S3::Bucket).to receive(:object).and_return(*s3_objects)
+        expect(s3_plugin).to receive(:process_log).at_least(size).and_call_original
+        expect(s3_plugin).to receive(:stop?).and_return(false).at_least(size)
+        expect(s3_plugin).to receive(:download_remote_file).and_return(true).at_least(size)
+        expect(s3_plugin).to receive(:process_local_log).and_return(true).at_least(size)
+
+        s3_plugin.process_files(queue)
+        expect(s3_plugin.send(:sincedb).read).to eq(s3_summary[0].last_modified)
+      end
+    end
   end
 end

data/spec/integration/s3_spec.rb

@@ -10,6 +10,7 @@ describe LogStash::Inputs::S3, :integration => true, :s3 => true do
 
     upload_file('../fixtures/uncompressed.log' , "#{prefix}uncompressed_1.log")
     upload_file('../fixtures/compressed.log.gz', "#{prefix}compressed_1.log.gz")
+    sleep(LogStash::Inputs::S3::CUTOFF_SECOND + 1)
   end
 
   after do
@@ -28,6 +29,7 @@ describe LogStash::Inputs::S3, :integration => true, :s3 => true do
                                "prefix" => prefix,
                                "temporary_directory" => temporary_directory } }
   let(:backup_prefix) { "backup/" }
+  let(:backup_bucket) { "logstash-s3-input-backup" }
 
   it "support prefix to scope the remote files" do
     events = fetch_events(minimal_settings)
@@ -49,13 +51,17 @@ describe LogStash::Inputs::S3, :integration => true, :s3 => true do
   end
 
   context "remote backup" do
+    before do
+      create_bucket(backup_bucket)
+    end
+
     it "another bucket" do
-      fetch_events(minimal_settings.merge({ "backup_to_bucket" => "logstash-s3-input-backup"}))
-      expect(list_remote_files("", "logstash-s3-input-backup").size).to eq(2)
+      fetch_events(minimal_settings.merge({ "backup_to_bucket" => backup_bucket}))
+      expect(list_remote_files("", backup_bucket).size).to eq(2)
     end
 
     after do
-      delete_bucket("logstash-s3-input-backup")
+      delete_bucket(backup_bucket)
     end
   end
 end

data/spec/support/helpers.rb

@@ -23,6 +23,10 @@ def list_remote_files(prefix, target_bucket = ENV['AWS_LOGSTASH_TEST_BUCKET'])
   bucket.objects(:prefix => prefix).collect(&:key)
 end
 
+def create_bucket(name)
+  s3object.bucket(name).create
+end
+
 def delete_bucket(name)
   s3object.bucket(name).objects.map(&:delete)
   s3object.bucket(name).delete
@@ -33,13 +37,16 @@ def s3object
 end
 
 class TestInfiniteS3Object
+  def initialize(s3_obj)
+    @s3_obj = s3_obj
+  end
+
   def each
     counter = 1
 
     loop do
-      yield "awesome-#{counter}"
+      yield @s3_obj
       counter +=1
     end
   end
-end
-
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: logstash-input-s3
 version: !ruby/object:Gem::Version
-  version: 3.5.0
+  version: 3.8.1
 platform: ruby
 authors:
 - Elastic
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2020-03-19 00:00:00.000000000 Z
+date: 2021-10-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
@@ -100,6 +100,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
+  name: logstash-mixin-ecs_compatibility_support
+  prerelease: false
+  type: :runtime
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
 description: This gem is a Logstash plugin required to be installed on top of the
   Logstash core pipeline using $LS_HOME/bin/logstash-plugin install gemname. This
   gem is not a stand-alone program
@@ -153,8 +167,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project:
-rubygems_version: 2.6.13
+rubygems_version: 3.1.6
 signing_key:
 specification_version: 4
 summary: Streams events from files in a S3 bucket