logstash-integration-jdbc 5.2.3 → 5.2.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2119dbab43322bdc083cf11602e69e63bcb1a957b537f5f44cfcd2b611df1893
-  data.tar.gz: 651e016e43bb17db730cc95ad0fd44bed483b9edfa4bbe86e52ced95947bb4a4
+  metadata.gz: c9cc41a07a8cb2f7e4b9a6993dd4d0cd69761b3fa95ba81238d0dd6267367311
+  data.tar.gz: 32866a340d87d91734586b3324a1349437abc624deb6ff9883ab998b0a4524a6
 SHA512:
-  metadata.gz: 0a90d6aa88365c9ddfe60bba20f6f06633733af4b910b75becc31c37a9e1a224d3aae8d5e62f03f8ae2a3ffd8a24d1b0635307aa152a836c18435a4b36ca1d5d
-  data.tar.gz: f6a6648d071bff9069f54448cfed03327cb0f5bb8c036d8b0c6e5adb949c69a75f923541fe6a1237eeb3e6fe35611ad0afd4f99e356e8dd42b1e2b19534bc5b7
+  metadata.gz: b3513ba2d9ade38a829ec0b0c47ad6c0978d0256e03fb4ac26e136a064db65b08a19b7ee8dd7dabe26851739877b82f0f9a550b2328473b8ffd62f194d569ebb
+  data.tar.gz: bf71a208f07bc00c213724cbcf175a9ede2f87bf4eaf85a926ee8e9f87b322e2dd79164db1db84ec60633a0b1cfa961e8fc6d3889c8ca25ea220fa437285af5d

data/CHANGELOG.md

@@ -1,3 +1,12 @@
+## 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) 
+
 ## 5.2.3
   - Performance: avoid contention on scheduler execution [#103](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/103)
 

data/Gemfile

@@ -9,3 +9,5 @@ if Dir.exist?(logstash_path) && use_logstash_source
   gem 'logstash-core', :path => "#{logstash_path}/logstash-core"
   gem 'logstash-core-plugin-api', :path => "#{logstash_path}/logstash-core-plugin-api"
 end
+
+gem 'rufus-scheduler', ENV['RUFUS_SCHEDULER_VERSION'] if ENV['RUFUS_SCHEDULER_VERSION']

data/lib/logstash/filters/jdbc/loader_schedule.rb

@@ -4,60 +4,26 @@ require "rufus/scheduler"
 
 module LogStash module Filters module Jdbc
   class LoaderSchedule < Validatable
-    attr_reader :schedule_frequency, :loader_schedule
-
-    def to_log_string
-      message = ""
-      message.concat "these months in the year [#{@cronline.months.to_a.join(", ")}];" unless @cronline.months.nil?
-      message.concat "these days in the month [#{@cronline.days.to_a.join(", ")}];" unless @cronline.days.nil?
-      message.concat "these hours in the day [#{@cronline.hours.to_a.join(", ")}];" unless @cronline.hours.nil?
-      message.concat "these minutes in the hour [#{@cronline.minutes.to_a.join(", ")}];" unless @cronline.minutes.nil?
-      message.concat "these seconds in the minute [#{@cronline.seconds.to_a.join(", ")}]" unless @cronline.seconds.nil?
-      if !message.empty?
-        message.prepend "Scheduled for: "
-      end
-      message
-    end
+    attr_reader :loader_schedule
 
     private
 
-    def post_initialize
-      if valid?
-        # From the Rufus::Scheduler docs:
-        #   By default, rufus-scheduler sleeps 0.300 second between every step.
-        #   At each step it checks for jobs to trigger and so on.
-        # set the frequency to 2.5 seconds if we are not reloading in the seconds timeframe
-        # rufus scheduler thread should respond to stop quickly enough.
-        if only_seconds_set?
-          @schedule_frequency = 0.3
-        else
-          @schedule_frequency = 2.5
-        end
-      end
-    end
-
-
-    def only_seconds_set?
-      @cronline.seconds &&
-        @cronline.minutes.nil? &&
-        @cronline.hours.nil? &&
-        @cronline.days.nil? &&
-        @cronline.months.nil?
-    end
-
+    # @overload
     def parse_options
       @loader_schedule = @options
 
-      unless @loader_schedule.is_a?(String)
+      if @loader_schedule.is_a?(String)
+        begin
+          # Rufus::Scheduler 3.0 - 3.6 methods signature: parse_cron(o, opts)
+          # since Rufus::Scheduler 3.7 methods signature: parse_cron(o, opts={})
+          @cronline = Rufus::Scheduler.parse_cron(@loader_schedule, {})
+        rescue => e
+          @option_errors << "The loader_schedule option is invalid: #{e.message}"
+        end
+      else
         @option_errors << "The loader_schedule option must be a string"
       end
 
-      begin
-        @cronline = Rufus::Scheduler::CronLine.new(@loader_schedule)
-      rescue => e
-        @option_errors << "The loader_schedule option is invalid: #{e.message}"
-      end
-
       @valid = @option_errors.empty?
     end
   end

data/lib/logstash/filters/jdbc/repeating_load_runner.rb

@@ -3,8 +3,6 @@ require_relative "single_load_runner"
 
 module LogStash module Filters module Jdbc
   class RepeatingLoadRunner < SingleLoadRunner
-   # info - attr_reader :local, :loaders, :preloaders
-
     def repeated_load
       local.repopulate_all(loaders)
       @reload_counter.increment

data/lib/logstash/filters/jdbc/single_load_runner.rb

@@ -26,10 +26,6 @@ module LogStash module Filters module Jdbc
     def repeated_load
     end
 
-    def call
-      repeated_load
-    end
-
     def reload_count
       @reload_counter.value
     end

data/lib/logstash/filters/jdbc_static.rb

@@ -3,6 +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_relative "jdbc/loader"
 require_relative "jdbc/loader_schedule"
 require_relative "jdbc/repeating_load_runner"
@@ -174,9 +175,23 @@ 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
 
@@ -191,17 +206,15 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
     @processor = Jdbc::LookupProcessor.new(@local_lookups, global_lookup_options)
     runner_args.unshift(@processor.local)
     if @loader_schedule
-      args = []
       @loader_runner = Jdbc::RepeatingLoadRunner.new(*runner_args)
       @loader_runner.initial_load
-      cronline = Jdbc::LoaderSchedule.new(@loader_schedule)
-      cronline.to_log_string.tap do |msg|
-        logger.info("Scheduler operations: #{msg}") unless msg.empty?
+      @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
-      logger.info("Scheduler scan for work frequency is: #{cronline.schedule_frequency}")
-      rufus_args = {:max_work_threads => 1, :frequency => cronline.schedule_frequency}
-      @scheduler = Rufus::Scheduler.new(rufus_args)
-      @scheduler.cron(cronline.loader_schedule, @loader_runner)
     else
       @loader_runner = Jdbc::SingleLoadRunner.new(*runner_args)
       @loader_runner.initial_load

data/lib/logstash/inputs/jdbc.rb

@@ -9,6 +9,7 @@ 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 "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"
@@ -178,8 +179,10 @@ 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
@@ -230,11 +233,33 @@ 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
+
+    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
+
     require "rufus/scheduler"
     prepare_jdbc_connection
 
@@ -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)
@@ -295,19 +320,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
     load_driver
     if @schedule
       # input thread (Java) name example "[my-oracle]<jdbc"
-      @scheduler = LogStash::PluginMixins::Jdbc::Scheduler.new(
-          :max_work_threads => 1,
-          :thread_name => "[#{id}]<jdbc__scheduler",
-          # 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
-          :frequency => 1.0,
-      )
-      @scheduler.schedule_cron @schedule do
-        execute_query(queue)
-      end
-
+      @scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
+          start_cron_scheduler(@schedule, thread_name: "[#{id}]<jdbc__scheduler") { execute_query(queue) }
       @scheduler.join
     else
       execute_query(queue)
@@ -374,4 +388,23 @@ 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/scheduler.rb

@@ -12,6 +12,34 @@ module LogStash module PluginMixins module Jdbc
     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

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

@@ -5,10 +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)
-      if plugin.record_last_run
-        handler = FileHandler.new(plugin.last_run_metadata_path)
-      end
+      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         = '5.2.3'
+  s.version         = '5.2.6'
   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"
@@ -34,7 +34,9 @@ Gem::Specification.new do |s|
 
   s.add_runtime_dependency 'tzinfo'
   s.add_runtime_dependency 'tzinfo-data'
-  s.add_runtime_dependency 'rufus-scheduler', '~> 3.0.9'
+  # 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'

data/spec/filters/jdbc/repeating_load_runner_spec.rb

@@ -17,7 +17,7 @@ module LogStash module Filters module Jdbc
         expect(local_db).to receive(:populate_all).once.with(loaders)
         expect(local_db).to receive(:repopulate_all).once.with(loaders)
         runner.initial_load
-        subject.call
+        subject.repeated_load
       end
     end
   end

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
@@ -1114,6 +1118,86 @@ 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 +1516,6 @@ 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) }

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: logstash-integration-jdbc
 version: !ruby/object:Gem::Version
-  version: 5.2.3
+  version: 5.2.6
 platform: ruby
 authors:
 - Elastic
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2022-02-16 00:00:00.000000000 Z
+date: 2022-05-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
@@ -131,17 +131,17 @@ dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 3.0.9
+        version: '0'
   name: rufus-scheduler
   prerelease: false
   type: :runtime
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 3.0.9
+        version: '0'
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement
     requirements:
@@ -275,7 +275,6 @@ 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

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