logstash-integration-jdbc 5.2.4 → 5.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: ad77fc92ab07e1da7e4c6285075b6583f9f293c073199b614e3852cd22bbc6a9
-  data.tar.gz: 9e7b2c0f0c16930903bdee1f22b2651db7cf9f80d01a12f212b7a40155d4ebea
+  metadata.gz: d9937a8c1c5ab26acb4cd887f4bdb9dfff51e9348c7187ef2d44ff6055d137ee
+  data.tar.gz: 0042c3e29b5b08abe8baa4f84e4c240960bbc615778df9c77415f8db7985c831
 SHA512:
-  metadata.gz: 2928a01bbe1ef418f36d6ddc7847416fc6d43278b20242926037bef02e8a429ba1d4587f1bd88f8d52de2ec1fef26e749aeb358de82a8bf69c36e5c3c9d42f8e
-  data.tar.gz: 8aae596826ec1d9155e0f8bdfad9a882e92fa2be610f89d7c67a7b3f44f0e6937bcbafd78900457d92b20a5a28188163f7b4f25040d8deb229f505ab6faf2c22
+  metadata.gz: dbc38871c106b045a500645cf6c4517b30ce15e52bfc68520845e22276efdb4a21df3e4a7709ae682cd6b55ff76e063a615b9fe397795272535aa7c5298a9a92
+  data.tar.gz: 70b8e1dc84ba60d478affb22c93400cb153d4d0adc4904667970220ac03cb9d22b94e28f1b1a5504a30b8592c543d3234ee441d0ed57a31aefef70dfb6c2b9a3

data/CHANGELOG.md

@@ -1,3 +1,12 @@
+## 5.3.0
+  - Refactor: start using scheduler mixin [#110](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/110)
+
+## 5.2.6
+  - Fix: change default path of 'last_run_metadata_path' to be rooted in the LS data.path folder and not in $HOME [#106](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/106)
+
+## 5.2.5
+  - Fix: do not execute more queries with debug logging [#109](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/109)
+
 ## 5.2.4
   - Fix: compatibility with all (>= 3.0) rufus-scheduler versions [#97](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/97) 
 

data/lib/logstash/filters/jdbc_static.rb

@@ -3,7 +3,7 @@ require "logstash-integration-jdbc_jars"
 require "logstash/filters/base"
 require "logstash/namespace"
 require "logstash/plugin_mixins/ecs_compatibility_support"
-require "logstash/plugin_mixins/jdbc/scheduler"
+require "logstash/plugin_mixins/scheduler"
 require_relative "jdbc/loader"
 require_relative "jdbc/loader_schedule"
 require_relative "jdbc/repeating_load_runner"
@@ -19,6 +19,8 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
   # adds ecs_compatibility config which could be :disabled or :v1
   include LogStash::PluginMixins::ECSCompatibilitySupport(:disabled, :v1, :v8 => :v1)
 
+  include LogStash::PluginMixins::Scheduler
+
   config_name "jdbc_static"
 
   # Define the loaders, an Array of Hashes, to fetch remote data and create local tables.
@@ -162,7 +164,6 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
   end
 
   def close
-    @scheduler.stop if @scheduler
     @parsed_loaders.each(&:close)
     @processor.close
   end
@@ -175,9 +176,23 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
   private
 
   def prepare_data_dir
+    # cleanup existing Derby file left behind in $HOME
+    derby_log = "#{ENV['HOME']}/derby.log"
+    if ::File.exist?(derby_log)
+      begin
+        ::File.delete(derby_log)
+      rescue Errno::EPERM => e
+        @logger.warn("Can't delete temporary file #{derby_log} due to access permissions")
+      rescue e
+        @logger.warn("Can't delete temporary file #{derby_log}", {message => e.message})
+      end
+    end
+
     # later, when local persistent databases are allowed set this property to LS_HOME/data/jdbc-static/
     # must take multi-pipelines into account and more than one config using the same jdbc-static settings
-    java.lang.System.setProperty("derby.system.home", ENV["HOME"])
+    path_data = Pathname.new(LogStash::SETTINGS.get_value("path.data")).join("plugins", "shared", "derby_home")
+    path_data.mkpath
+    java.lang.System.setProperty("derby.system.home", path_data.to_path)
     logger.info("derby.system.home is: #{java.lang.System.getProperty("derby.system.home")}")
   end
 
@@ -194,13 +209,7 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
     if @loader_schedule
       @loader_runner = Jdbc::RepeatingLoadRunner.new(*runner_args)
       @loader_runner.initial_load
-      @scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
-          start_cron_scheduler(@loader_schedule, thread_name: "[#{id}]-jdbc_static__scheduler") { @loader_runner.repeated_load }
-      cron_job = @scheduler.cron_jobs.first
-      if cron_job
-        frequency = cron_job.respond_to?(:rough_frequency) ? cron_job.rough_frequency : cron_job.frequency
-        logger.info("Loaders will execute every #{frequency} seconds", loader_schedule: @loader_schedule)
-      end
+      scheduler.cron(@loader_schedule) { @loader_runner.repeated_load }
     else
       @loader_runner = Jdbc::SingleLoadRunner.new(*runner_args)
       @loader_runner.initial_load

data/lib/logstash/inputs/jdbc.rb

@@ -3,12 +3,12 @@ require "logstash/inputs/base"
 require "logstash/namespace"
 require "logstash/plugin_mixins/jdbc/common"
 require "logstash/plugin_mixins/jdbc/jdbc"
-require "logstash/plugin_mixins/jdbc/scheduler"
 require "logstash/plugin_mixins/ecs_compatibility_support"
 require "logstash/plugin_mixins/ecs_compatibility_support/target_check"
 require "logstash/plugin_mixins/validator_support/field_reference_validation_adapter"
-
 require "logstash/plugin_mixins/event_support/event_factory_adapter"
+require "logstash/plugin_mixins/scheduler"
+require "fileutils"
 
 # this require_relative returns early unless the JRuby version is between 9.2.0.0 and 9.2.8.0
 require_relative "tzinfo_jruby_patch"
@@ -146,6 +146,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
   # adds :field_reference validator adapter
   extend LogStash::PluginMixins::ValidatorSupport::FieldReferenceValidationAdapter
 
+  include LogStash::PluginMixins::Scheduler
+
   config_name "jdbc"
 
   # If undefined, Logstash will complain, even if codec is unused.
@@ -178,8 +180,10 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
   # exactly once.
   config :schedule, :validate => :string
 
-  # Path to file with last run time
-  config :last_run_metadata_path, :validate => :string, :default => "#{ENV['HOME']}/.logstash_jdbc_last_run"
+  # Path to file with last run time.
+  # The default will write file to `<path.data>/plugins/inputs/jdbc/logstash_jdbc_last_run`
+  # NOTE: it must be a file path and not a directory path
+  config :last_run_metadata_path, :validate => :string
 
   # Use an incremental column value rather than a timestamp
   config :use_column_value, :validate => :boolean, :default => false
@@ -230,12 +234,33 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
   config :target, :validate => :field_reference, :required => false
 
   attr_reader :database # for test mocking/stubbing
+  attr_reader :last_run_metadata_file_path # path to the file used as last run storage
 
   public
 
   def register
     @logger = self.logger
-    require "rufus/scheduler"
+
+    if @record_last_run
+      if @last_run_metadata_path.nil?
+        logstash_data_path = LogStash::SETTINGS.get_value("path.data")
+        logstash_data_path = Pathname.new(logstash_data_path).join("plugins", "inputs", "jdbc")
+        # Ensure that the filepath exists before writing, since it's deeply nested.
+        logstash_data_path.mkpath
+        logstash_data_file_path = logstash_data_path.join("logstash_jdbc_last_run")
+
+        ensure_default_metadatafile_location(logstash_data_file_path)
+
+        @last_run_metadata_file_path = logstash_data_file_path.to_path
+      else
+        #  validate the path is a file and not a directory
+        if Pathname.new(@last_run_metadata_path).directory?
+          raise LogStash::ConfigurationError.new("The \"last_run_metadata_path\" argument must point to a file, received a directory: \"#{last_run_metadata_path}\"")
+        end
+        @last_run_metadata_file_path = @last_run_metadata_path
+      end
+    end
+
     prepare_jdbc_connection
 
     if @use_column_value
@@ -259,8 +284,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
       end
     end
 
-    set_value_tracker(LogStash::PluginMixins::Jdbc::ValueTracking.build_last_value_tracker(self))
-    set_statement_logger(LogStash::PluginMixins::Jdbc::CheckedCountLogger.new(@logger))
+    set_value_tracker LogStash::PluginMixins::Jdbc::ValueTracking.build_last_value_tracker(self)
+    set_statement_handler LogStash::PluginMixins::Jdbc::StatementHandler.build_statement_handler(self)
 
     @enable_encoding = !@charset.nil? || !@columns_charset.empty?
 
@@ -283,8 +308,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
   end # def register
 
   # test injection points
-  def set_statement_logger(instance)
-    @statement_handler = LogStash::PluginMixins::Jdbc::StatementHandler.build_statement_handler(self, instance)
+  def set_statement_handler(handler)
+    @statement_handler = handler
   end
 
   def set_value_tracker(instance)
@@ -294,22 +319,16 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
   def run(queue)
     load_driver
     if @schedule
-      # input thread (Java) name example "[my-oracle]<jdbc"
-      @scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
-          start_cron_scheduler(@schedule, thread_name: "[#{id}]<jdbc__scheduler") { execute_query(queue) }
-      @scheduler.join
+      # scheduler input thread name example: "[my-oracle]|input|jdbc|scheduler"
+      scheduler.cron(@schedule) { execute_query(queue) }
+      scheduler.join
     else
       execute_query(queue)
     end
   end # def run
 
-  def close
-    @scheduler.shutdown if @scheduler
-  end
-
   def stop
     close_jdbc_connection
-    @scheduler.shutdown(:wait) if @scheduler
   end
 
   private
@@ -363,4 +382,23 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
       value
     end
   end
+
+  def ensure_default_metadatafile_location(metadata_new_path)
+    old_default_path = Pathname.new("#{ENV['HOME']}/.logstash_jdbc_last_run")
+
+    if old_default_path.exist? && !metadata_new_path.exist?
+      # Previous versions of the plugin hosted the state file into $HOME/.logstash_jdbc_last_run.
+      # Copy in the new location
+      FileUtils.cp(old_default_path.to_path, metadata_new_path.to_path)
+      begin
+        # If there is a permission error in the delete of the old file inform the user to give
+        # the correct access rights
+        ::File.delete(old_default_path.to_path)
+        @logger.info("Successfully moved the #{old_default_path.to_path} into #{metadata_new_path.to_path}")
+      rescue e
+        @logger.warn("Using new metadata file at #{metadata_new_path.to_path} but #{old_default_path} can't be removed.")
+      end
+    end
+  end
+
 end end end # class LogStash::Inputs::Jdbc

data/lib/logstash/plugin_mixins/jdbc/jdbc.rb

@@ -4,7 +4,6 @@ require "logstash/config/mixin"
 require "time"
 require "date"
 require_relative "value_tracking"
-require_relative "checked_count_logger"
 require_relative "statement_handler"
 
 java_import java.util.concurrent.locks.ReentrantLock

data/lib/logstash/plugin_mixins/jdbc/statement_handler.rb

@@ -2,7 +2,7 @@
 
 module LogStash module PluginMixins module Jdbc
   class StatementHandler
-    def self.build_statement_handler(plugin, logger)
+    def self.build_statement_handler(plugin)
       if plugin.use_prepared_statements
         klass = PreparedStatementHandler
       else
@@ -16,27 +16,39 @@ module LogStash module PluginMixins module Jdbc
           klass = NormalStatementHandler
         end
       end
-      klass.new(plugin, logger)
+      klass.new(plugin)
     end
 
-    attr_reader :statement, :parameters, :statement_logger
+    attr_reader :statement, :parameters
 
-    def initialize(plugin, statement_logger)
+    def initialize(plugin)
       @statement = plugin.statement
-      @statement_logger = statement_logger
-      post_init(plugin)
     end
 
     def build_query(db, sql_last_value)
-      # override in subclass
+      fail NotImplementedError # override in subclass
     end
 
-    def post_init(plugin)
-      # override in subclass, if needed
-    end
   end
 
   class NormalStatementHandler < StatementHandler
+
+    attr_reader :parameters
+
+    def initialize(plugin)
+      super(plugin)
+      @parameter_keys = ["sql_last_value"] + plugin.parameters.keys
+      @parameters = plugin.parameters.inject({}) do |hash,(k,v)|
+        case v
+        when LogStash::Timestamp
+          hash[k.to_sym] = v.time
+        else
+          hash[k.to_sym] = v
+        end
+        hash
+      end
+    end
+
     # Performs the query, yielding once per row of data
     # @param db [Sequel::Database]
     # @param sql_last_value [Integer|DateTime|Time]
@@ -52,27 +64,18 @@ module LogStash module PluginMixins module Jdbc
 
     def build_query(db, sql_last_value)
       parameters[:sql_last_value] = sql_last_value
-      query = db[statement, parameters]
-      statement_logger.log_statement_parameters(statement, parameters, query)
-      query
+      db[statement, parameters]
     end
 
-    def post_init(plugin)
-      @parameter_keys = ["sql_last_value"] + plugin.parameters.keys
-      @parameters = plugin.parameters.inject({}) do |hash,(k,v)|
-        case v
-        when LogStash::Timestamp
-          hash[k.to_sym] = v.time
-        else
-          hash[k.to_sym] = v
-        end
-        hash
-      end
-    end
   end
 
   class PagedNormalStatementHandler < NormalStatementHandler
-    attr_reader :jdbc_page_size
+
+    def initialize(plugin)
+      super(plugin)
+      @jdbc_page_size = plugin.jdbc_page_size
+      @logger = plugin.logger
+    end
 
     # Performs the query, respecting our pagination settings, yielding once per row of data
     # @param db [Sequel::Database]
@@ -81,16 +84,22 @@ module LogStash module PluginMixins module Jdbc
     def perform_query(db, sql_last_value)
       query = build_query(db, sql_last_value)
       query.each_page(@jdbc_page_size) do |paged_dataset|
+        log_dataset_page(paged_dataset) if @logger.debug?
         paged_dataset.each do |row|
           yield row
         end
       end
     end
 
-    def post_init(plugin)
-      super(plugin)
-      @jdbc_page_size = plugin.jdbc_page_size
+    private
+
+    # @param paged_dataset [Sequel::Dataset::Pagination] like object
+    def log_dataset_page(paged_dataset)
+      @logger.debug "fetching paged dataset", current_page: paged_dataset.current_page,
+                                              record_count: paged_dataset.current_page_record_count,
+                                              total_record_count: paged_dataset.pagination_record_count
     end
+
   end
 
   class ExplicitPagingModeStatementHandler < PagedNormalStatementHandler
@@ -101,20 +110,29 @@ module LogStash module PluginMixins module Jdbc
     def perform_query(db, sql_last_value)
       query = build_query(db, sql_last_value)
       offset = 0
+      page_size = @jdbc_page_size
       loop do
         rows_in_page = 0
-        query.with_sql(query.sql, offset: offset, size: jdbc_page_size).each do |row|
+        query.with_sql(query.sql, offset: offset, size: page_size).each do |row|
           yield row
           rows_in_page += 1
         end
-        break unless rows_in_page == jdbc_page_size
-        offset += jdbc_page_size
+        break unless rows_in_page == page_size
+        offset += page_size
       end
     end
   end
 
   class PreparedStatementHandler < StatementHandler
-    attr_reader :name, :bind_values_array, :statement_prepared, :prepared
+    attr_reader :name, :bind_values_array, :statement_prepared, :prepared, :parameters
+
+    def initialize(plugin)
+      super(plugin)
+      @name = plugin.prepared_statement_name.to_sym
+      @bind_values_array = plugin.prepared_statement_bind_values
+      @parameters = plugin.parameters
+      @statement_prepared = Concurrent::AtomicBoolean.new(false)
+    end
 
     # Performs the query, ignoring our pagination settings, yielding once per row of data
     # @param db [Sequel::Database]
@@ -142,7 +160,6 @@ module LogStash module PluginMixins module Jdbc
         db.set_prepared_statement(name, prepared)
       end
       bind_value_sql_last_value(sql_last_value)
-      statement_logger.log_statement_parameters(statement, parameters, nil)
       begin
         db.call(name, parameters)
       rescue => e
@@ -153,17 +170,6 @@ module LogStash module PluginMixins module Jdbc
       end
     end
 
-    def post_init(plugin)
-      # don't log statement count when using prepared statements for now...
-      # needs enhancement to allow user to supply a bindable count prepared statement in settings.
-      @statement_logger.disable_count
-
-      @name = plugin.prepared_statement_name.to_sym
-      @bind_values_array = plugin.prepared_statement_bind_values
-      @parameters = plugin.parameters
-      @statement_prepared = Concurrent::AtomicBoolean.new(false)
-    end
-
     def create_bind_values_hash
       hash = {}
       bind_values_array.each_with_index {|v,i| hash[:"p#{i}"] = v}

data/lib/logstash/plugin_mixins/jdbc/value_tracking.rb

@@ -5,7 +5,7 @@ module LogStash module PluginMixins module Jdbc
   class ValueTracking
 
     def self.build_last_value_tracker(plugin)
-      handler = plugin.record_last_run ? FileHandler.new(plugin.last_run_metadata_path) : NullFileHandler.new(plugin.last_run_metadata_path)
+      handler = plugin.record_last_run ? FileHandler.new(plugin.last_run_metadata_file_path) : NullFileHandler.new(plugin.last_run_metadata_file_path)
       if plugin.clean_run
         handler.clean
       end

data/logstash-integration-jdbc.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name = 'logstash-integration-jdbc'
-  s.version         = '5.2.4'
+  s.version         = '5.3.0'
   s.licenses = ['Apache License (2.0)']
   s.summary         = "Integration with JDBC - input and filter plugins"
   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"
@@ -34,12 +34,11 @@ Gem::Specification.new do |s|
 
   s.add_runtime_dependency 'tzinfo'
   s.add_runtime_dependency 'tzinfo-data'
-  # plugin maintains compatibility with < 3.5 (3.0.9)
-  # but works with newer rufus-scheduler >= 3.5 as well
-  s.add_runtime_dependency 'rufus-scheduler'
+
   s.add_runtime_dependency 'logstash-mixin-ecs_compatibility_support', '~>1.3'
   s.add_runtime_dependency "logstash-mixin-validator_support", '~> 1.0'
   s.add_runtime_dependency "logstash-mixin-event_support", '~> 1.0'
+  s.add_runtime_dependency "logstash-mixin-scheduler", '~> 1.0'
 
   s.add_development_dependency "childprocess"
   s.add_development_dependency 'logstash-devutils', '>= 2.3'

data/spec/filters/jdbc_static_spec.rb

@@ -5,6 +5,7 @@ require "sequel"
 require "sequel/adapters/jdbc"
 require "stud/temporary"
 require "timecop"
+require "pathname"
 
 # LogStash::Logging::Logger::configure_logging("WARN")
 
@@ -85,6 +86,15 @@ module LogStash module Filters
 
     let(:ipaddr) { ".3.1.1" }
 
+    describe "verify derby path property" do
+      it "should be set into Logstash data path" do
+        plugin.register
+
+        expected = Pathname.new(LogStash::SETTINGS.get_value("path.data")).join("plugins", "shared", "derby_home").to_path
+        expect(java.lang.System.getProperty("derby.system.home")).to eq(expected)
+      end
+    end
+
     describe "non scheduled operation" do
       after { plugin.close }
 

data/spec/inputs/integration/integ_spec.rb

@@ -41,6 +41,31 @@ describe LogStash::Inputs::Jdbc, :integration => true do
       expect(event.get('first_name')).to eq("Mark")
       expect(event.get('last_name')).to eq("Guckenheimer")
     end
+
+    context 'with paging' do
+      let(:settings) do
+        super().merge 'jdbc_paging_enabled' => true, 'jdbc_page_size' => 1,
+                      "statement" => 'SELECT * FROM "employee" WHERE EMP_NO >= :p1 ORDER BY EMP_NO',
+                      'parameters' => { 'p1' => 0 }
+      end
+
+      before do # change plugin logger level to debug - to exercise logging
+        logger = plugin.class.name.gsub('::', '.').downcase
+        logger = org.apache.logging.log4j.LogManager.getLogger(logger)
+        @prev_logger_level = [ logger.getName, logger.getLevel ]
+        org.apache.logging.log4j.core.config.Configurator.setLevel logger.getName, org.apache.logging.log4j.Level::DEBUG
+      end
+
+      after do
+        org.apache.logging.log4j.core.config.Configurator.setLevel *@prev_logger_level
+      end
+
+      it "should populate the event with database entries" do
+        plugin.run(queue)
+        event = queue.pop
+        expect(event.get('first_name')).to eq('David')
+      end
+    end
   end
 
   context "when supplying a non-existent library" do

data/spec/inputs/jdbc_spec.rb

@@ -9,6 +9,7 @@ require "timecop"
 require "stud/temporary"
 require "time"
 require "date"
+require "pathname"
 
 # We do not need to set TZ env var anymore because we can have 'Sequel.application_timezone' set to utc by default now.
 
@@ -51,6 +52,9 @@ describe LogStash::Inputs::Jdbc do
       db.drop_table(:types_table)
       db.drop_table(:test1_table)
     end
+
+    last_run_default_path = LogStash::SETTINGS.get_value("path.data")
+    FileUtils.rm_f("#{last_run_default_path}/plugins/inputs/jdbc/logstash_jdbc_last_run")
   end
 
   context "when registering and tearing down" do
@@ -247,18 +251,6 @@ describe LogStash::Inputs::Jdbc do
       Timecop.return
     end
 
-    it "cleans up scheduler resources on close" do
-      runner = Thread.new do
-        plugin.run(queue)
-      end
-      sleep 1
-      plugin.do_close
-
-      scheduler = plugin.instance_variable_get(:@scheduler)
-      expect(scheduler).to_not be_nil
-      expect(scheduler.down?).to be_truthy
-    end
-
   end
 
   context "when scheduling and previous runs are to be preserved" do
@@ -1114,6 +1106,86 @@ describe LogStash::Inputs::Jdbc do
     end
   end
 
+  context "when state is persisted" do
+    context "to file" do
+      let(:settings) do
+        {
+          "statement" => "SELECT * FROM test_table",
+          "record_last_run" => true
+        }
+      end
+
+      before do
+        plugin.register
+      end
+
+      after do
+        plugin.stop
+      end
+
+      context "with default last_run_metadata_path" do
+        it "should save state in data.data subpath" do
+          path = LogStash::SETTINGS.get_value("path.data")
+          expect(plugin.last_run_metadata_file_path).to start_with(path)
+        end
+      end
+
+      context "with customized last_run_metadata_path" do
+        let(:settings) { super().merge({ "last_run_metadata_path" => Stud::Temporary.pathname })}
+
+        it "should save state in data.data subpath" do
+          expect(plugin.last_run_metadata_file_path).to start_with(settings["last_run_metadata_path"])
+        end
+      end
+    end
+
+    context "with customized last_run_metadata_path point to directory" do
+      let(:settings) do
+        path = Stud::Temporary.pathname
+        Pathname.new(path).tap {|path| path.mkpath}
+        super().merge({ "last_run_metadata_path" => path})
+      end
+
+      it "raise configuration error" do
+        expect { plugin.register }.to raise_error(LogStash::ConfigurationError)
+      end
+    end
+  end
+
+  context "update the previous default last_run_metadata_path" do
+    let(:settings) do
+      {
+        "statement" => "SELECT * FROM test_table",
+        "record_last_run" => true
+      }
+    end
+
+    let(:fake_home) do
+       path = Stud::Temporary.pathname
+       Pathname.new(path).tap {|path| path.mkpath}
+       path
+    end
+
+    context "when a file exists" do
+      before do
+        # in a faked HOME folder save a valid previous last_run metadata file
+        allow(ENV).to receive(:[]).with('HOME').and_return(fake_home)
+        File.open("#{ENV['HOME']}/.logstash_jdbc_last_run", 'w') do |file|
+          file.write("--- !ruby/object:DateTime '2022-03-08 08:10:00.486889000 Z'")
+        end
+      end
+
+      it "should be moved" do
+        plugin.register
+
+        expect(::File.exist?("#{ENV['HOME']}/.logstash_jdbc_last_run")).to be false
+        path = LogStash::SETTINGS.get_value("path.data")
+        full_path = "#{path}/plugins/inputs/jdbc/logstash_jdbc_last_run"
+        expect(::File.exist?(full_path)).to be true
+      end
+    end
+  end
+
   context "when setting fetch size" do
 
     let(:settings) do
@@ -1432,54 +1504,6 @@ describe LogStash::Inputs::Jdbc do
     end
   end
 
-  context "when debug logging and a count query raises a count related error" do
-    let(:settings) do
-      { "statement" => "SELECT * from types_table" }
-    end
-    let(:logger) { double("logger", :debug? => true) }
-    let(:statement_logger) { LogStash::PluginMixins::Jdbc::CheckedCountLogger.new(logger) }
-    let(:value_tracker) { double("value tracker", :set_value => nil, :write => nil) }
-    let(:msg) { 'Java::JavaSql::SQLSyntaxErrorException: Dynamic SQL Error; SQL error code = -104; Token unknown - line 1, column 105; LIMIT [SQLState:42000, ISC error code:335544634]' }
-    let(:error_args) do
-      {"exception" => msg}
-    end
-
-    before do
-      db << "INSERT INTO types_table (num, string, started_at, custom_time, ranking) VALUES (1, 'A test', '1999-12-31', '1999-12-31 23:59:59', 95.67)"
-      plugin.register
-      plugin.set_statement_logger(statement_logger)
-      plugin.set_value_tracker(value_tracker)
-      allow(value_tracker).to receive(:value).and_return("bar")
-      allow(statement_logger).to receive(:execute_count).once.and_raise(StandardError.new(msg))
-    end
-
-    after do
-      plugin.stop
-    end
-
-    context "if the count query raises an error" do
-      it "should log a debug line without a count key as its unknown whether a count works at this stage" do
-        expect(logger).to receive(:warn).once.with("Attempting a count query raised an error, the generated count statement is most likely incorrect but check networking, authentication or your statement syntax", error_args)
-        expect(logger).to receive(:warn).once.with("Ongoing count statement generation is being prevented")
-        expect(logger).to receive(:debug).once.with("Executing JDBC query", :statement => settings["statement"], :parameters => {:sql_last_value=>"bar"})
-        plugin.run(queue)
-        queue.pop
-      end
-
-      it "should create an event normally" do
-        allow(logger).to receive(:warn)
-        allow(logger).to receive(:debug)
-        plugin.run(queue)
-        event = queue.pop
-        expect(event.get("num")).to eq(1)
-        expect(event.get("string")).to eq("A test")
-        expect(event.get("started_at")).to be_a_logstash_timestamp_equivalent_to("1999-12-31T00:00:00.000Z")
-        expect(event.get("custom_time")).to be_a_logstash_timestamp_equivalent_to("1999-12-31T23:59:59.000Z")
-        expect(event.get("ranking").to_f).to eq(95.67)
-      end
-    end
-  end
-
   context "when an unreadable jdbc_driver_path entry is present" do
     let(:driver_jar_path) do
       jar_file = $CLASSPATH.find { |name| name.index(Jdbc::Derby.driver_jar) }
@@ -1681,7 +1705,7 @@ describe LogStash::Inputs::Jdbc do
       let(:jdbc_driver_class) { "org.apache.NonExistentDriver" }
       it "raises a loading error" do
         expect { plugin.send(:load_driver) }.to raise_error LogStash::PluginLoadingError,
-                                                            /java.lang.ClassNotFoundException: org.apache.NonExistentDriver/
+                                                            /ClassNotFoundException: org.apache.NonExistentDriver/
       end
     end
   end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: logstash-integration-jdbc
 version: !ruby/object:Gem::Version
-  version: 5.2.4
+  version: 5.3.0
 platform: ruby
 authors:
 - Elastic
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-03-14 00:00:00.000000000 Z
+date: 2022-06-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
@@ -131,38 +131,38 @@ dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
-  name: rufus-scheduler
+        version: '1.3'
+  name: logstash-mixin-ecs_compatibility_support
   prerelease: false
   type: :runtime
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '1.3'
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.3'
-  name: logstash-mixin-ecs_compatibility_support
+        version: '1.0'
+  name: logstash-mixin-validator_support
   prerelease: false
   type: :runtime
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '1.0'
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '1.0'
-  name: logstash-mixin-validator_support
+  name: logstash-mixin-event_support
   prerelease: false
   type: :runtime
   version_requirements: !ruby/object:Gem::Requirement
@@ -176,7 +176,7 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '1.0'
-  name: logstash-mixin-event_support
+  name: logstash-mixin-scheduler
   prerelease: false
   type: :runtime
   version_requirements: !ruby/object:Gem::Requirement
@@ -275,10 +275,8 @@ files:
 - lib/logstash/filters/jdbc_streaming.rb
 - lib/logstash/inputs/jdbc.rb
 - lib/logstash/inputs/tzinfo_jruby_patch.rb
-- lib/logstash/plugin_mixins/jdbc/checked_count_logger.rb
 - lib/logstash/plugin_mixins/jdbc/common.rb
 - lib/logstash/plugin_mixins/jdbc/jdbc.rb
-- lib/logstash/plugin_mixins/jdbc/scheduler.rb
 - lib/logstash/plugin_mixins/jdbc/statement_handler.rb
 - lib/logstash/plugin_mixins/jdbc/value_tracking.rb
 - lib/logstash/plugin_mixins/jdbc_streaming.rb
@@ -307,7 +305,6 @@ files:
 - spec/helpers/derbyrun.jar
 - spec/inputs/integration/integ_spec.rb
 - spec/inputs/jdbc_spec.rb
-- spec/plugin_mixins/jdbc/scheduler_spec.rb
 - spec/plugin_mixins/jdbc_streaming/parameter_handler_spec.rb
 - vendor/jar-dependencies/org/apache/derby/derby/10.14.1.0/derby-10.14.1.0.jar
 - vendor/jar-dependencies/org/apache/derby/derbyclient/10.14.1.0/derbyclient-10.14.1.0.jar
@@ -360,5 +357,4 @@ test_files:
 - spec/helpers/derbyrun.jar
 - spec/inputs/integration/integ_spec.rb
 - spec/inputs/jdbc_spec.rb
-- spec/plugin_mixins/jdbc/scheduler_spec.rb
 - spec/plugin_mixins/jdbc_streaming/parameter_handler_spec.rb

data/lib/logstash/plugin_mixins/jdbc/checked_count_logger.rb

@@ -1,43 +0,0 @@
-# encoding: utf-8
-
-module LogStash module PluginMixins module Jdbc
-  class CheckedCountLogger
-    def initialize(logger)
-      @logger = logger
-      @needs_check = true
-      @count_is_supported = false
-      @in_debug = @logger.debug?
-    end
-
-    def disable_count
-      @needs_check = false
-      @count_is_supported = false
-    end
-
-    def log_statement_parameters(statement, parameters, query)
-      return unless @in_debug
-      check_count_query(query) if @needs_check && query
-      if @count_is_supported
-        @logger.debug("Executing JDBC query", :statement => statement, :parameters => parameters, :count => execute_count(query))
-      else
-        @logger.debug("Executing JDBC query", :statement => statement, :parameters => parameters)
-      end
-    end
-
-    def check_count_query(query)
-      @needs_check = false
-      begin
-        execute_count(query)
-        @count_is_supported = true
-      rescue Exception => e
-        @logger.warn("Attempting a count query raised an error, the generated count statement is most likely incorrect but check networking, authentication or your statement syntax", "exception" => e.message)
-        @logger.warn("Ongoing count statement generation is being prevented")
-        @count_is_supported = false
-      end
-    end
-
-    def execute_count(query)
-      query.count
-    end
-  end
-end end end

data/lib/logstash/plugin_mixins/jdbc/scheduler.rb

@@ -1,175 +0,0 @@
-require 'rufus/scheduler'
-
-require 'logstash/util/loggable'
-
-module LogStash module PluginMixins module Jdbc
-  class Scheduler < Rufus::Scheduler
-
-    include LogStash::Util::Loggable
-
-    # Rufus::Scheduler >= 3.4 moved the Time impl into a gem EoTime = ::EtOrbi::EoTime`
-    # Rufus::Scheduler 3.1 - 3.3 using it's own Time impl `Rufus::Scheduler::ZoTime`
-    TimeImpl = defined?(Rufus::Scheduler::EoTime) ? Rufus::Scheduler::EoTime :
-                   (defined?(Rufus::Scheduler::ZoTime) ? Rufus::Scheduler::ZoTime : ::Time)
-
-    # @param cron [String] cron-line
-    # @param opts [Hash] scheduler options
-    # @return scheduler instance
-    def self.start_cron_scheduler(cron, opts = {}, &block)
-      unless block_given?
-        raise ArgumentError, 'missing (cron scheduler) block - worker task to execute'
-      end
-      scheduler = new_scheduler(opts)
-      scheduler.schedule_cron(cron, &block)
-      scheduler
-    end
-
-    # @param opts [Hash] scheduler options
-    # @return scheduler instance
-    def self.new_scheduler(opts)
-      unless opts.key?(:thread_name)
-        raise ArgumentError, 'thread_name: option is required to be able to distinguish multiple scheduler threads'
-      end
-      opts[:max_work_threads] ||= 1
-      # amount the scheduler thread sleeps between checking whether jobs
-      # should trigger, default is 0.3 which is a bit too often ...
-      # in theory the cron expression '* * * * * *' supports running jobs
-      # every second but this is very rare, we could potentially go higher
-      opts[:frequency] ||= 1.0
-
-      new(opts)
-    end
-
-    # @overload
-    def timeout_jobs
-      # Rufus relies on `Thread.list` which is a blocking operation and with many schedulers
-      # (and threads) within LS will have a negative impact on performance as scheduler
-      # threads will end up waiting to obtain the `Thread.list` lock.
-      #
-      # However, this isn't necessary we can easily detect whether there are any jobs
-      # that might need to timeout: only when `@opts[:timeout]` is set causes worker thread(s)
-      # to have a `Thread.current[:rufus_scheduler_timeout]` that is not nil
-      return unless @opts[:timeout]
-      super
-    end
-
-    # @overload
-    def work_threads(query = :all)
-      if query == :__all_no_cache__ # special case from JobDecorator#start_work_thread
-        @_work_threads = nil # when a new worker thread is being added reset
-        return super(:all)
-      end
-
-      # Gets executed every time a job is triggered, we're going to cache the
-      # worker threads for this scheduler (to avoid `Thread.list`) - they only
-      # change when a new thread is being started from #start_work_thread ...
-      work_threads = @_work_threads
-      if work_threads.nil?
-        work_threads = threads.select { |t| t[:rufus_scheduler_work_thread] }
-        @_work_threads = work_threads
-      end
-
-      case query
-      when :active then work_threads.select { |t| t[:rufus_scheduler_job] }
-      when :vacant then work_threads.reject { |t| t[:rufus_scheduler_job] }
-      else work_threads
-      end
-    end
-
-    # @overload
-    def on_error(job, err)
-      details = { exception: err.class, message: err.message, backtrace: err.backtrace }
-      details[:cause] = err.cause if err.cause
-
-      details[:now] = debug_format_time(TimeImpl.now)
-      details[:last_time] = (debug_format_time(job.last_time) rescue nil)
-      details[:next_time] = (debug_format_time(job.next_time) rescue nil)
-      details[:job] = job
-
-      details[:opts] = @opts
-      details[:started_at] = started_at
-      details[:thread] = thread.inspect
-      details[:jobs_size] = @jobs.size
-      details[:work_threads_size] = work_threads.size
-      details[:work_queue_size] = work_queue.size
-
-      logger.error("Scheduler intercepted an error:", details)
-
-    rescue => e
-      logger.error("Scheduler failed in #on_error #{e.inspect}")
-    end
-
-    def debug_format_time(time)
-      # EtOrbi::EoTime used by (newer) Rufus::Scheduler has to_debug_s https://git.io/JyiPj
-      time.respond_to?(:to_debug_s) ? time.to_debug_s : time.strftime("%Y-%m-%dT%H:%M:%S.%L")
-    end
-    private :debug_format_time
-
-    # @private helper used by JobDecorator
-    def work_thread_name_prefix
-      ( @opts[:thread_name] || "#{@thread_key}_scheduler" ) + '_worker-'
-    end
-
-    protected
-
-    # @overload
-    def start
-      ret = super() # @thread[:name] = @opts[:thread_name] || "#{@thread_key}_scheduler"
-
-      # at least set thread.name for easier thread dump analysis
-      if @thread.is_a?(Thread) && @thread.respond_to?(:name=)
-        @thread.name = @thread[:name] if @thread[:name]
-      end
-
-      ret
-    end
-
-    # @overload
-    def do_schedule(job_type, t, callable, opts, return_job_instance, block)
-      job_or_id = super
-
-      job_or_id.extend JobDecorator if return_job_instance
-
-      job_or_id
-    end
-
-    module JobDecorator
-
-      def start_work_thread
-        prev_thread_count = @scheduler.work_threads.size
-
-        ret = super() # does not return Thread instance in 3.0
-
-        work_threads = @scheduler.work_threads(:__all_no_cache__)
-        while prev_thread_count == work_threads.size # very unlikely
-          Thread.pass
-          work_threads = @scheduler.work_threads(:__all_no_cache__)
-        end
-
-        work_thread_name_prefix = @scheduler.work_thread_name_prefix
-
-        work_threads.sort! do |t1, t2|
-          if t1[:name].nil?
-            t2[:name].nil? ? 0 : +1 # nils at the end
-          elsif t2[:name].nil?
-            t1[:name].nil? ? 0 : -1
-          else
-            t1[:name] <=> t2[:name]
-          end
-        end
-
-        work_threads.each_with_index do |thread, i|
-          unless thread[:name]
-            thread[:name] = "#{work_thread_name_prefix}#{sprintf('%02i', i)}"
-            thread.name = thread[:name] if thread.respond_to?(:name=)
-            # e.g. "[oracle]<jdbc_scheduler_worker-00"
-          end
-        end
-
-        ret
-      end
-
-    end
-
-  end
-end end end

data/spec/plugin_mixins/jdbc/scheduler_spec.rb

@@ -1,78 +0,0 @@
-# encoding: utf-8
-require "logstash/devutils/rspec/spec_helper"
-require "logstash/plugin_mixins/jdbc/scheduler"
-
-describe LogStash::PluginMixins::Jdbc::Scheduler do
-
-  let(:thread_name) { '[test]<jdbc_scheduler' }
-
-  let(:opts) do
-    { :max_work_threads => 2, :thread_name => thread_name }
-  end
-
-  subject(:scheduler) { LogStash::PluginMixins::Jdbc::Scheduler.new(opts) }
-
-  after { scheduler.stop(:wait) }
-
-  it "sets scheduler thread name" do
-    expect( scheduler.thread.name ).to include thread_name
-  end
-
-  context 'cron schedule' do
-
-    before do
-      scheduler.schedule_cron('* * * * * *') { sleep 1.25 } # every second
-    end
-
-    it "sets worker thread names" do
-      sleep 3.0
-      threads = scheduler.work_threads
-      threads.sort! { |t1, t2| (t1.name || '') <=> (t2.name || '') }
-
-      expect( threads.size ).to eql 2
-      expect( threads.first.name ).to eql "#{thread_name}_worker-00"
-      expect( threads.last.name ).to eql "#{thread_name}_worker-01"
-    end
-
-  end
-
-  context 'every 1s' do
-
-    before do
-      scheduler.schedule_in('1s') { raise 'TEST' } # every second
-    end
-
-    it "logs errors handled" do
-      expect( scheduler.logger ).to receive(:error).with /Scheduler intercepted an error/, hash_including(:message => 'TEST')
-      sleep 1.5
-    end
-
-  end
-
-  context 'work threads' do
-
-    let(:opts) { super().merge :max_work_threads => 3 }
-
-    let(:counter) { java.util.concurrent.atomic.AtomicLong.new(0) }
-
-    before do
-      scheduler.schedule_cron('* * * * * *') { counter.increment_and_get; sleep 3.25 } # every second
-    end
-
-    it "are working" do
-      sleep(0.05) while counter.get == 0
-      expect( scheduler.work_threads.size ).to eql 1
-      sleep(0.05) while counter.get == 1
-      expect( scheduler.work_threads.size ).to eql 2
-      sleep(0.05) while counter.get == 2
-      expect( scheduler.work_threads.size ).to eql 3
-
-      sleep 1.25
-      expect( scheduler.work_threads.size ).to eql 3
-      sleep 1.25
-      expect( scheduler.work_threads.size ).to eql 3
-    end
-
-  end
-
-end