logstash-integration-jdbc 5.2.3 → 5.2.6
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +9 -0
- data/Gemfile +2 -0
- data/lib/logstash/filters/jdbc/loader_schedule.rb +11 -45
- data/lib/logstash/filters/jdbc/repeating_load_runner.rb +0 -2
- data/lib/logstash/filters/jdbc/single_load_runner.rb +0 -4
- data/lib/logstash/filters/jdbc_static.rb +22 -9
- data/lib/logstash/inputs/jdbc.rb +52 -19
- data/lib/logstash/plugin_mixins/jdbc/jdbc.rb +0 -1
- data/lib/logstash/plugin_mixins/jdbc/scheduler.rb +28 -0
- data/lib/logstash/plugin_mixins/jdbc/statement_handler.rb +51 -45
- data/lib/logstash/plugin_mixins/jdbc/value_tracking.rb +1 -4
- data/logstash-integration-jdbc.gemspec +4 -2
- data/spec/filters/jdbc/repeating_load_runner_spec.rb +1 -1
- data/spec/filters/jdbc_static_spec.rb +10 -0
- data/spec/inputs/integration/integ_spec.rb +25 -0
- data/spec/inputs/jdbc_spec.rb +84 -48
- metadata +6 -7
- data/lib/logstash/plugin_mixins/jdbc/checked_count_logger.rb +0 -43
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: c9cc41a07a8cb2f7e4b9a6993dd4d0cd69761b3fa95ba81238d0dd6267367311
|
4
|
+
data.tar.gz: 32866a340d87d91734586b3324a1349437abc624deb6ff9883ab998b0a4524a6
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: b3513ba2d9ade38a829ec0b0c47ad6c0978d0256e03fb4ac26e136a064db65b08a19b7ee8dd7dabe26851739877b82f0f9a550b2328473b8ffd62f194d569ebb
|
7
|
+
data.tar.gz: bf71a208f07bc00c213724cbcf175a9ede2f87bf4eaf85a926ee8e9f87b322e2dd79164db1db84ec60633a0b1cfa961e8fc6d3889c8ca25ea220fa437285af5d
|
data/CHANGELOG.md
CHANGED
@@ -1,3 +1,12 @@
|
|
1
|
+
## 5.2.6
|
2
|
+
- 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)
|
3
|
+
|
4
|
+
## 5.2.5
|
5
|
+
- Fix: do not execute more queries with debug logging [#109](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/109)
|
6
|
+
|
7
|
+
## 5.2.4
|
8
|
+
- Fix: compatibility with all (>= 3.0) rufus-scheduler versions [#97](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/97)
|
9
|
+
|
1
10
|
## 5.2.3
|
2
11
|
- Performance: avoid contention on scheduler execution [#103](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/103)
|
3
12
|
|
data/Gemfile
CHANGED
@@ -9,3 +9,5 @@ if Dir.exist?(logstash_path) && use_logstash_source
|
|
9
9
|
gem 'logstash-core', :path => "#{logstash_path}/logstash-core"
|
10
10
|
gem 'logstash-core-plugin-api', :path => "#{logstash_path}/logstash-core-plugin-api"
|
11
11
|
end
|
12
|
+
|
13
|
+
gem 'rufus-scheduler', ENV['RUFUS_SCHEDULER_VERSION'] if ENV['RUFUS_SCHEDULER_VERSION']
|
@@ -4,60 +4,26 @@ require "rufus/scheduler"
|
|
4
4
|
|
5
5
|
module LogStash module Filters module Jdbc
|
6
6
|
class LoaderSchedule < Validatable
|
7
|
-
attr_reader :
|
8
|
-
|
9
|
-
def to_log_string
|
10
|
-
message = ""
|
11
|
-
message.concat "these months in the year [#{@cronline.months.to_a.join(", ")}];" unless @cronline.months.nil?
|
12
|
-
message.concat "these days in the month [#{@cronline.days.to_a.join(", ")}];" unless @cronline.days.nil?
|
13
|
-
message.concat "these hours in the day [#{@cronline.hours.to_a.join(", ")}];" unless @cronline.hours.nil?
|
14
|
-
message.concat "these minutes in the hour [#{@cronline.minutes.to_a.join(", ")}];" unless @cronline.minutes.nil?
|
15
|
-
message.concat "these seconds in the minute [#{@cronline.seconds.to_a.join(", ")}]" unless @cronline.seconds.nil?
|
16
|
-
if !message.empty?
|
17
|
-
message.prepend "Scheduled for: "
|
18
|
-
end
|
19
|
-
message
|
20
|
-
end
|
7
|
+
attr_reader :loader_schedule
|
21
8
|
|
22
9
|
private
|
23
10
|
|
24
|
-
|
25
|
-
if valid?
|
26
|
-
# From the Rufus::Scheduler docs:
|
27
|
-
# By default, rufus-scheduler sleeps 0.300 second between every step.
|
28
|
-
# At each step it checks for jobs to trigger and so on.
|
29
|
-
# set the frequency to 2.5 seconds if we are not reloading in the seconds timeframe
|
30
|
-
# rufus scheduler thread should respond to stop quickly enough.
|
31
|
-
if only_seconds_set?
|
32
|
-
@schedule_frequency = 0.3
|
33
|
-
else
|
34
|
-
@schedule_frequency = 2.5
|
35
|
-
end
|
36
|
-
end
|
37
|
-
end
|
38
|
-
|
39
|
-
|
40
|
-
def only_seconds_set?
|
41
|
-
@cronline.seconds &&
|
42
|
-
@cronline.minutes.nil? &&
|
43
|
-
@cronline.hours.nil? &&
|
44
|
-
@cronline.days.nil? &&
|
45
|
-
@cronline.months.nil?
|
46
|
-
end
|
47
|
-
|
11
|
+
# @overload
|
48
12
|
def parse_options
|
49
13
|
@loader_schedule = @options
|
50
14
|
|
51
|
-
|
15
|
+
if @loader_schedule.is_a?(String)
|
16
|
+
begin
|
17
|
+
# Rufus::Scheduler 3.0 - 3.6 methods signature: parse_cron(o, opts)
|
18
|
+
# since Rufus::Scheduler 3.7 methods signature: parse_cron(o, opts={})
|
19
|
+
@cronline = Rufus::Scheduler.parse_cron(@loader_schedule, {})
|
20
|
+
rescue => e
|
21
|
+
@option_errors << "The loader_schedule option is invalid: #{e.message}"
|
22
|
+
end
|
23
|
+
else
|
52
24
|
@option_errors << "The loader_schedule option must be a string"
|
53
25
|
end
|
54
26
|
|
55
|
-
begin
|
56
|
-
@cronline = Rufus::Scheduler::CronLine.new(@loader_schedule)
|
57
|
-
rescue => e
|
58
|
-
@option_errors << "The loader_schedule option is invalid: #{e.message}"
|
59
|
-
end
|
60
|
-
|
61
27
|
@valid = @option_errors.empty?
|
62
28
|
end
|
63
29
|
end
|
@@ -3,6 +3,7 @@ require "logstash-integration-jdbc_jars"
|
|
3
3
|
require "logstash/filters/base"
|
4
4
|
require "logstash/namespace"
|
5
5
|
require "logstash/plugin_mixins/ecs_compatibility_support"
|
6
|
+
require "logstash/plugin_mixins/jdbc/scheduler"
|
6
7
|
require_relative "jdbc/loader"
|
7
8
|
require_relative "jdbc/loader_schedule"
|
8
9
|
require_relative "jdbc/repeating_load_runner"
|
@@ -174,9 +175,23 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
|
|
174
175
|
private
|
175
176
|
|
176
177
|
def prepare_data_dir
|
178
|
+
# cleanup existing Derby file left behind in $HOME
|
179
|
+
derby_log = "#{ENV['HOME']}/derby.log"
|
180
|
+
if ::File.exist?(derby_log)
|
181
|
+
begin
|
182
|
+
::File.delete(derby_log)
|
183
|
+
rescue Errno::EPERM => e
|
184
|
+
@logger.warn("Can't delete temporary file #{derby_log} due to access permissions")
|
185
|
+
rescue e
|
186
|
+
@logger.warn("Can't delete temporary file #{derby_log}", {message => e.message})
|
187
|
+
end
|
188
|
+
end
|
189
|
+
|
177
190
|
# later, when local persistent databases are allowed set this property to LS_HOME/data/jdbc-static/
|
178
191
|
# must take multi-pipelines into account and more than one config using the same jdbc-static settings
|
179
|
-
|
192
|
+
path_data = Pathname.new(LogStash::SETTINGS.get_value("path.data")).join("plugins", "shared", "derby_home")
|
193
|
+
path_data.mkpath
|
194
|
+
java.lang.System.setProperty("derby.system.home", path_data.to_path)
|
180
195
|
logger.info("derby.system.home is: #{java.lang.System.getProperty("derby.system.home")}")
|
181
196
|
end
|
182
197
|
|
@@ -191,17 +206,15 @@ module LogStash module Filters class JdbcStatic < LogStash::Filters::Base
|
|
191
206
|
@processor = Jdbc::LookupProcessor.new(@local_lookups, global_lookup_options)
|
192
207
|
runner_args.unshift(@processor.local)
|
193
208
|
if @loader_schedule
|
194
|
-
args = []
|
195
209
|
@loader_runner = Jdbc::RepeatingLoadRunner.new(*runner_args)
|
196
210
|
@loader_runner.initial_load
|
197
|
-
|
198
|
-
|
199
|
-
|
211
|
+
@scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
|
212
|
+
start_cron_scheduler(@loader_schedule, thread_name: "[#{id}]-jdbc_static__scheduler") { @loader_runner.repeated_load }
|
213
|
+
cron_job = @scheduler.cron_jobs.first
|
214
|
+
if cron_job
|
215
|
+
frequency = cron_job.respond_to?(:rough_frequency) ? cron_job.rough_frequency : cron_job.frequency
|
216
|
+
logger.info("Loaders will execute every #{frequency} seconds", loader_schedule: @loader_schedule)
|
200
217
|
end
|
201
|
-
logger.info("Scheduler scan for work frequency is: #{cronline.schedule_frequency}")
|
202
|
-
rufus_args = {:max_work_threads => 1, :frequency => cronline.schedule_frequency}
|
203
|
-
@scheduler = Rufus::Scheduler.new(rufus_args)
|
204
|
-
@scheduler.cron(cronline.loader_schedule, @loader_runner)
|
205
218
|
else
|
206
219
|
@loader_runner = Jdbc::SingleLoadRunner.new(*runner_args)
|
207
220
|
@loader_runner.initial_load
|
data/lib/logstash/inputs/jdbc.rb
CHANGED
@@ -9,6 +9,7 @@ require "logstash/plugin_mixins/ecs_compatibility_support/target_check"
|
|
9
9
|
require "logstash/plugin_mixins/validator_support/field_reference_validation_adapter"
|
10
10
|
|
11
11
|
require "logstash/plugin_mixins/event_support/event_factory_adapter"
|
12
|
+
require "fileutils"
|
12
13
|
|
13
14
|
# this require_relative returns early unless the JRuby version is between 9.2.0.0 and 9.2.8.0
|
14
15
|
require_relative "tzinfo_jruby_patch"
|
@@ -178,8 +179,10 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
178
179
|
# exactly once.
|
179
180
|
config :schedule, :validate => :string
|
180
181
|
|
181
|
-
# Path to file with last run time
|
182
|
-
|
182
|
+
# Path to file with last run time.
|
183
|
+
# The default will write file to `<path.data>/plugins/inputs/jdbc/logstash_jdbc_last_run`
|
184
|
+
# NOTE: it must be a file path and not a directory path
|
185
|
+
config :last_run_metadata_path, :validate => :string
|
183
186
|
|
184
187
|
# Use an incremental column value rather than a timestamp
|
185
188
|
config :use_column_value, :validate => :boolean, :default => false
|
@@ -230,11 +233,33 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
230
233
|
config :target, :validate => :field_reference, :required => false
|
231
234
|
|
232
235
|
attr_reader :database # for test mocking/stubbing
|
236
|
+
attr_reader :last_run_metadata_file_path # path to the file used as last run storage
|
233
237
|
|
234
238
|
public
|
235
239
|
|
236
240
|
def register
|
237
241
|
@logger = self.logger
|
242
|
+
|
243
|
+
if @record_last_run
|
244
|
+
if @last_run_metadata_path.nil?
|
245
|
+
logstash_data_path = LogStash::SETTINGS.get_value("path.data")
|
246
|
+
logstash_data_path = Pathname.new(logstash_data_path).join("plugins", "inputs", "jdbc")
|
247
|
+
# Ensure that the filepath exists before writing, since it's deeply nested.
|
248
|
+
logstash_data_path.mkpath
|
249
|
+
logstash_data_file_path = logstash_data_path.join("logstash_jdbc_last_run")
|
250
|
+
|
251
|
+
ensure_default_metadatafile_location(logstash_data_file_path)
|
252
|
+
|
253
|
+
@last_run_metadata_file_path = logstash_data_file_path.to_path
|
254
|
+
else
|
255
|
+
# validate the path is a file and not a directory
|
256
|
+
if Pathname.new(@last_run_metadata_path).directory?
|
257
|
+
raise LogStash::ConfigurationError.new("The \"last_run_metadata_path\" argument must point to a file, received a directory: \"#{last_run_metadata_path}\"")
|
258
|
+
end
|
259
|
+
@last_run_metadata_file_path = @last_run_metadata_path
|
260
|
+
end
|
261
|
+
end
|
262
|
+
|
238
263
|
require "rufus/scheduler"
|
239
264
|
prepare_jdbc_connection
|
240
265
|
|
@@ -259,8 +284,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
259
284
|
end
|
260
285
|
end
|
261
286
|
|
262
|
-
set_value_tracker
|
263
|
-
|
287
|
+
set_value_tracker LogStash::PluginMixins::Jdbc::ValueTracking.build_last_value_tracker(self)
|
288
|
+
set_statement_handler LogStash::PluginMixins::Jdbc::StatementHandler.build_statement_handler(self)
|
264
289
|
|
265
290
|
@enable_encoding = !@charset.nil? || !@columns_charset.empty?
|
266
291
|
|
@@ -283,8 +308,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
283
308
|
end # def register
|
284
309
|
|
285
310
|
# test injection points
|
286
|
-
def
|
287
|
-
@statement_handler =
|
311
|
+
def set_statement_handler(handler)
|
312
|
+
@statement_handler = handler
|
288
313
|
end
|
289
314
|
|
290
315
|
def set_value_tracker(instance)
|
@@ -295,19 +320,8 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
295
320
|
load_driver
|
296
321
|
if @schedule
|
297
322
|
# input thread (Java) name example "[my-oracle]<jdbc"
|
298
|
-
@scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
|
299
|
-
:
|
300
|
-
:thread_name => "[#{id}]<jdbc__scheduler",
|
301
|
-
# amount the scheduler thread sleeps between checking whether jobs
|
302
|
-
# should trigger, default is 0.3 which is a bit too often ...
|
303
|
-
# in theory the cron expression '* * * * * *' supports running jobs
|
304
|
-
# every second but this is very rare, we could potentially go higher
|
305
|
-
:frequency => 1.0,
|
306
|
-
)
|
307
|
-
@scheduler.schedule_cron @schedule do
|
308
|
-
execute_query(queue)
|
309
|
-
end
|
310
|
-
|
323
|
+
@scheduler = LogStash::PluginMixins::Jdbc::Scheduler.
|
324
|
+
start_cron_scheduler(@schedule, thread_name: "[#{id}]<jdbc__scheduler") { execute_query(queue) }
|
311
325
|
@scheduler.join
|
312
326
|
else
|
313
327
|
execute_query(queue)
|
@@ -374,4 +388,23 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
374
388
|
value
|
375
389
|
end
|
376
390
|
end
|
391
|
+
|
392
|
+
def ensure_default_metadatafile_location(metadata_new_path)
|
393
|
+
old_default_path = Pathname.new("#{ENV['HOME']}/.logstash_jdbc_last_run")
|
394
|
+
|
395
|
+
if old_default_path.exist? && !metadata_new_path.exist?
|
396
|
+
# Previous versions of the plugin hosted the state file into $HOME/.logstash_jdbc_last_run.
|
397
|
+
# Copy in the new location
|
398
|
+
FileUtils.cp(old_default_path.to_path, metadata_new_path.to_path)
|
399
|
+
begin
|
400
|
+
# If there is a permission error in the delete of the old file inform the user to give
|
401
|
+
# the correct access rights
|
402
|
+
::File.delete(old_default_path.to_path)
|
403
|
+
@logger.info("Successfully moved the #{old_default_path.to_path} into #{metadata_new_path.to_path}")
|
404
|
+
rescue e
|
405
|
+
@logger.warn("Using new metadata file at #{metadata_new_path.to_path} but #{old_default_path} can't be removed.")
|
406
|
+
end
|
407
|
+
end
|
408
|
+
end
|
409
|
+
|
377
410
|
end end end # class LogStash::Inputs::Jdbc
|
@@ -12,6 +12,34 @@ module LogStash module PluginMixins module Jdbc
|
|
12
12
|
TimeImpl = defined?(Rufus::Scheduler::EoTime) ? Rufus::Scheduler::EoTime :
|
13
13
|
(defined?(Rufus::Scheduler::ZoTime) ? Rufus::Scheduler::ZoTime : ::Time)
|
14
14
|
|
15
|
+
# @param cron [String] cron-line
|
16
|
+
# @param opts [Hash] scheduler options
|
17
|
+
# @return scheduler instance
|
18
|
+
def self.start_cron_scheduler(cron, opts = {}, &block)
|
19
|
+
unless block_given?
|
20
|
+
raise ArgumentError, 'missing (cron scheduler) block - worker task to execute'
|
21
|
+
end
|
22
|
+
scheduler = new_scheduler(opts)
|
23
|
+
scheduler.schedule_cron(cron, &block)
|
24
|
+
scheduler
|
25
|
+
end
|
26
|
+
|
27
|
+
# @param opts [Hash] scheduler options
|
28
|
+
# @return scheduler instance
|
29
|
+
def self.new_scheduler(opts)
|
30
|
+
unless opts.key?(:thread_name)
|
31
|
+
raise ArgumentError, 'thread_name: option is required to be able to distinguish multiple scheduler threads'
|
32
|
+
end
|
33
|
+
opts[:max_work_threads] ||= 1
|
34
|
+
# amount the scheduler thread sleeps between checking whether jobs
|
35
|
+
# should trigger, default is 0.3 which is a bit too often ...
|
36
|
+
# in theory the cron expression '* * * * * *' supports running jobs
|
37
|
+
# every second but this is very rare, we could potentially go higher
|
38
|
+
opts[:frequency] ||= 1.0
|
39
|
+
|
40
|
+
new(opts)
|
41
|
+
end
|
42
|
+
|
15
43
|
# @overload
|
16
44
|
def timeout_jobs
|
17
45
|
# Rufus relies on `Thread.list` which is a blocking operation and with many schedulers
|
@@ -2,7 +2,7 @@
|
|
2
2
|
|
3
3
|
module LogStash module PluginMixins module Jdbc
|
4
4
|
class StatementHandler
|
5
|
-
def self.build_statement_handler(plugin
|
5
|
+
def self.build_statement_handler(plugin)
|
6
6
|
if plugin.use_prepared_statements
|
7
7
|
klass = PreparedStatementHandler
|
8
8
|
else
|
@@ -16,27 +16,39 @@ module LogStash module PluginMixins module Jdbc
|
|
16
16
|
klass = NormalStatementHandler
|
17
17
|
end
|
18
18
|
end
|
19
|
-
klass.new(plugin
|
19
|
+
klass.new(plugin)
|
20
20
|
end
|
21
21
|
|
22
|
-
attr_reader :statement, :parameters
|
22
|
+
attr_reader :statement, :parameters
|
23
23
|
|
24
|
-
def initialize(plugin
|
24
|
+
def initialize(plugin)
|
25
25
|
@statement = plugin.statement
|
26
|
-
@statement_logger = statement_logger
|
27
|
-
post_init(plugin)
|
28
26
|
end
|
29
27
|
|
30
28
|
def build_query(db, sql_last_value)
|
31
|
-
# override in subclass
|
29
|
+
fail NotImplementedError # override in subclass
|
32
30
|
end
|
33
31
|
|
34
|
-
def post_init(plugin)
|
35
|
-
# override in subclass, if needed
|
36
|
-
end
|
37
32
|
end
|
38
33
|
|
39
34
|
class NormalStatementHandler < StatementHandler
|
35
|
+
|
36
|
+
attr_reader :parameters
|
37
|
+
|
38
|
+
def initialize(plugin)
|
39
|
+
super(plugin)
|
40
|
+
@parameter_keys = ["sql_last_value"] + plugin.parameters.keys
|
41
|
+
@parameters = plugin.parameters.inject({}) do |hash,(k,v)|
|
42
|
+
case v
|
43
|
+
when LogStash::Timestamp
|
44
|
+
hash[k.to_sym] = v.time
|
45
|
+
else
|
46
|
+
hash[k.to_sym] = v
|
47
|
+
end
|
48
|
+
hash
|
49
|
+
end
|
50
|
+
end
|
51
|
+
|
40
52
|
# Performs the query, yielding once per row of data
|
41
53
|
# @param db [Sequel::Database]
|
42
54
|
# @param sql_last_value [Integer|DateTime|Time]
|
@@ -52,27 +64,18 @@ module LogStash module PluginMixins module Jdbc
|
|
52
64
|
|
53
65
|
def build_query(db, sql_last_value)
|
54
66
|
parameters[:sql_last_value] = sql_last_value
|
55
|
-
|
56
|
-
statement_logger.log_statement_parameters(statement, parameters, query)
|
57
|
-
query
|
67
|
+
db[statement, parameters]
|
58
68
|
end
|
59
69
|
|
60
|
-
def post_init(plugin)
|
61
|
-
@parameter_keys = ["sql_last_value"] + plugin.parameters.keys
|
62
|
-
@parameters = plugin.parameters.inject({}) do |hash,(k,v)|
|
63
|
-
case v
|
64
|
-
when LogStash::Timestamp
|
65
|
-
hash[k.to_sym] = v.time
|
66
|
-
else
|
67
|
-
hash[k.to_sym] = v
|
68
|
-
end
|
69
|
-
hash
|
70
|
-
end
|
71
|
-
end
|
72
70
|
end
|
73
71
|
|
74
72
|
class PagedNormalStatementHandler < NormalStatementHandler
|
75
|
-
|
73
|
+
|
74
|
+
def initialize(plugin)
|
75
|
+
super(plugin)
|
76
|
+
@jdbc_page_size = plugin.jdbc_page_size
|
77
|
+
@logger = plugin.logger
|
78
|
+
end
|
76
79
|
|
77
80
|
# Performs the query, respecting our pagination settings, yielding once per row of data
|
78
81
|
# @param db [Sequel::Database]
|
@@ -81,16 +84,22 @@ module LogStash module PluginMixins module Jdbc
|
|
81
84
|
def perform_query(db, sql_last_value)
|
82
85
|
query = build_query(db, sql_last_value)
|
83
86
|
query.each_page(@jdbc_page_size) do |paged_dataset|
|
87
|
+
log_dataset_page(paged_dataset) if @logger.debug?
|
84
88
|
paged_dataset.each do |row|
|
85
89
|
yield row
|
86
90
|
end
|
87
91
|
end
|
88
92
|
end
|
89
93
|
|
90
|
-
|
91
|
-
|
92
|
-
|
94
|
+
private
|
95
|
+
|
96
|
+
# @param paged_dataset [Sequel::Dataset::Pagination] like object
|
97
|
+
def log_dataset_page(paged_dataset)
|
98
|
+
@logger.debug "fetching paged dataset", current_page: paged_dataset.current_page,
|
99
|
+
record_count: paged_dataset.current_page_record_count,
|
100
|
+
total_record_count: paged_dataset.pagination_record_count
|
93
101
|
end
|
102
|
+
|
94
103
|
end
|
95
104
|
|
96
105
|
class ExplicitPagingModeStatementHandler < PagedNormalStatementHandler
|
@@ -101,20 +110,29 @@ module LogStash module PluginMixins module Jdbc
|
|
101
110
|
def perform_query(db, sql_last_value)
|
102
111
|
query = build_query(db, sql_last_value)
|
103
112
|
offset = 0
|
113
|
+
page_size = @jdbc_page_size
|
104
114
|
loop do
|
105
115
|
rows_in_page = 0
|
106
|
-
query.with_sql(query.sql, offset: offset, size:
|
116
|
+
query.with_sql(query.sql, offset: offset, size: page_size).each do |row|
|
107
117
|
yield row
|
108
118
|
rows_in_page += 1
|
109
119
|
end
|
110
|
-
break unless rows_in_page ==
|
111
|
-
offset +=
|
120
|
+
break unless rows_in_page == page_size
|
121
|
+
offset += page_size
|
112
122
|
end
|
113
123
|
end
|
114
124
|
end
|
115
125
|
|
116
126
|
class PreparedStatementHandler < StatementHandler
|
117
|
-
attr_reader :name, :bind_values_array, :statement_prepared, :prepared
|
127
|
+
attr_reader :name, :bind_values_array, :statement_prepared, :prepared, :parameters
|
128
|
+
|
129
|
+
def initialize(plugin)
|
130
|
+
super(plugin)
|
131
|
+
@name = plugin.prepared_statement_name.to_sym
|
132
|
+
@bind_values_array = plugin.prepared_statement_bind_values
|
133
|
+
@parameters = plugin.parameters
|
134
|
+
@statement_prepared = Concurrent::AtomicBoolean.new(false)
|
135
|
+
end
|
118
136
|
|
119
137
|
# Performs the query, ignoring our pagination settings, yielding once per row of data
|
120
138
|
# @param db [Sequel::Database]
|
@@ -142,7 +160,6 @@ module LogStash module PluginMixins module Jdbc
|
|
142
160
|
db.set_prepared_statement(name, prepared)
|
143
161
|
end
|
144
162
|
bind_value_sql_last_value(sql_last_value)
|
145
|
-
statement_logger.log_statement_parameters(statement, parameters, nil)
|
146
163
|
begin
|
147
164
|
db.call(name, parameters)
|
148
165
|
rescue => e
|
@@ -153,17 +170,6 @@ module LogStash module PluginMixins module Jdbc
|
|
153
170
|
end
|
154
171
|
end
|
155
172
|
|
156
|
-
def post_init(plugin)
|
157
|
-
# don't log statement count when using prepared statements for now...
|
158
|
-
# needs enhancement to allow user to supply a bindable count prepared statement in settings.
|
159
|
-
@statement_logger.disable_count
|
160
|
-
|
161
|
-
@name = plugin.prepared_statement_name.to_sym
|
162
|
-
@bind_values_array = plugin.prepared_statement_bind_values
|
163
|
-
@parameters = plugin.parameters
|
164
|
-
@statement_prepared = Concurrent::AtomicBoolean.new(false)
|
165
|
-
end
|
166
|
-
|
167
173
|
def create_bind_values_hash
|
168
174
|
hash = {}
|
169
175
|
bind_values_array.each_with_index {|v,i| hash[:"p#{i}"] = v}
|
@@ -5,10 +5,7 @@ module LogStash module PluginMixins module Jdbc
|
|
5
5
|
class ValueTracking
|
6
6
|
|
7
7
|
def self.build_last_value_tracker(plugin)
|
8
|
-
handler = plugin.record_last_run ? FileHandler.new(plugin.
|
9
|
-
if plugin.record_last_run
|
10
|
-
handler = FileHandler.new(plugin.last_run_metadata_path)
|
11
|
-
end
|
8
|
+
handler = plugin.record_last_run ? FileHandler.new(plugin.last_run_metadata_file_path) : NullFileHandler.new(plugin.last_run_metadata_file_path)
|
12
9
|
if plugin.clean_run
|
13
10
|
handler.clean
|
14
11
|
end
|
@@ -1,6 +1,6 @@
|
|
1
1
|
Gem::Specification.new do |s|
|
2
2
|
s.name = 'logstash-integration-jdbc'
|
3
|
-
s.version = '5.2.
|
3
|
+
s.version = '5.2.6'
|
4
4
|
s.licenses = ['Apache License (2.0)']
|
5
5
|
s.summary = "Integration with JDBC - input and filter plugins"
|
6
6
|
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|
|
|
34
34
|
|
35
35
|
s.add_runtime_dependency 'tzinfo'
|
36
36
|
s.add_runtime_dependency 'tzinfo-data'
|
37
|
-
|
37
|
+
# plugin maintains compatibility with < 3.5 (3.0.9)
|
38
|
+
# but works with newer rufus-scheduler >= 3.5 as well
|
39
|
+
s.add_runtime_dependency 'rufus-scheduler'
|
38
40
|
s.add_runtime_dependency 'logstash-mixin-ecs_compatibility_support', '~>1.3'
|
39
41
|
s.add_runtime_dependency "logstash-mixin-validator_support", '~> 1.0'
|
40
42
|
s.add_runtime_dependency "logstash-mixin-event_support", '~> 1.0'
|
@@ -5,6 +5,7 @@ require "sequel"
|
|
5
5
|
require "sequel/adapters/jdbc"
|
6
6
|
require "stud/temporary"
|
7
7
|
require "timecop"
|
8
|
+
require "pathname"
|
8
9
|
|
9
10
|
# LogStash::Logging::Logger::configure_logging("WARN")
|
10
11
|
|
@@ -85,6 +86,15 @@ module LogStash module Filters
|
|
85
86
|
|
86
87
|
let(:ipaddr) { ".3.1.1" }
|
87
88
|
|
89
|
+
describe "verify derby path property" do
|
90
|
+
it "should be set into Logstash data path" do
|
91
|
+
plugin.register
|
92
|
+
|
93
|
+
expected = Pathname.new(LogStash::SETTINGS.get_value("path.data")).join("plugins", "shared", "derby_home").to_path
|
94
|
+
expect(java.lang.System.getProperty("derby.system.home")).to eq(expected)
|
95
|
+
end
|
96
|
+
end
|
97
|
+
|
88
98
|
describe "non scheduled operation" do
|
89
99
|
after { plugin.close }
|
90
100
|
|
@@ -41,6 +41,31 @@ describe LogStash::Inputs::Jdbc, :integration => true do
|
|
41
41
|
expect(event.get('first_name')).to eq("Mark")
|
42
42
|
expect(event.get('last_name')).to eq("Guckenheimer")
|
43
43
|
end
|
44
|
+
|
45
|
+
context 'with paging' do
|
46
|
+
let(:settings) do
|
47
|
+
super().merge 'jdbc_paging_enabled' => true, 'jdbc_page_size' => 1,
|
48
|
+
"statement" => 'SELECT * FROM "employee" WHERE EMP_NO >= :p1 ORDER BY EMP_NO',
|
49
|
+
'parameters' => { 'p1' => 0 }
|
50
|
+
end
|
51
|
+
|
52
|
+
before do # change plugin logger level to debug - to exercise logging
|
53
|
+
logger = plugin.class.name.gsub('::', '.').downcase
|
54
|
+
logger = org.apache.logging.log4j.LogManager.getLogger(logger)
|
55
|
+
@prev_logger_level = [ logger.getName, logger.getLevel ]
|
56
|
+
org.apache.logging.log4j.core.config.Configurator.setLevel logger.getName, org.apache.logging.log4j.Level::DEBUG
|
57
|
+
end
|
58
|
+
|
59
|
+
after do
|
60
|
+
org.apache.logging.log4j.core.config.Configurator.setLevel *@prev_logger_level
|
61
|
+
end
|
62
|
+
|
63
|
+
it "should populate the event with database entries" do
|
64
|
+
plugin.run(queue)
|
65
|
+
event = queue.pop
|
66
|
+
expect(event.get('first_name')).to eq('David')
|
67
|
+
end
|
68
|
+
end
|
44
69
|
end
|
45
70
|
|
46
71
|
context "when supplying a non-existent library" do
|
data/spec/inputs/jdbc_spec.rb
CHANGED
@@ -9,6 +9,7 @@ require "timecop"
|
|
9
9
|
require "stud/temporary"
|
10
10
|
require "time"
|
11
11
|
require "date"
|
12
|
+
require "pathname"
|
12
13
|
|
13
14
|
# We do not need to set TZ env var anymore because we can have 'Sequel.application_timezone' set to utc by default now.
|
14
15
|
|
@@ -51,6 +52,9 @@ describe LogStash::Inputs::Jdbc do
|
|
51
52
|
db.drop_table(:types_table)
|
52
53
|
db.drop_table(:test1_table)
|
53
54
|
end
|
55
|
+
|
56
|
+
last_run_default_path = LogStash::SETTINGS.get_value("path.data")
|
57
|
+
FileUtils.rm_f("#{last_run_default_path}/plugins/inputs/jdbc/logstash_jdbc_last_run")
|
54
58
|
end
|
55
59
|
|
56
60
|
context "when registering and tearing down" do
|
@@ -1114,6 +1118,86 @@ describe LogStash::Inputs::Jdbc do
|
|
1114
1118
|
end
|
1115
1119
|
end
|
1116
1120
|
|
1121
|
+
context "when state is persisted" do
|
1122
|
+
context "to file" do
|
1123
|
+
let(:settings) do
|
1124
|
+
{
|
1125
|
+
"statement" => "SELECT * FROM test_table",
|
1126
|
+
"record_last_run" => true
|
1127
|
+
}
|
1128
|
+
end
|
1129
|
+
|
1130
|
+
before do
|
1131
|
+
plugin.register
|
1132
|
+
end
|
1133
|
+
|
1134
|
+
after do
|
1135
|
+
plugin.stop
|
1136
|
+
end
|
1137
|
+
|
1138
|
+
context "with default last_run_metadata_path" do
|
1139
|
+
it "should save state in data.data subpath" do
|
1140
|
+
path = LogStash::SETTINGS.get_value("path.data")
|
1141
|
+
expect(plugin.last_run_metadata_file_path).to start_with(path)
|
1142
|
+
end
|
1143
|
+
end
|
1144
|
+
|
1145
|
+
context "with customized last_run_metadata_path" do
|
1146
|
+
let(:settings) { super().merge({ "last_run_metadata_path" => Stud::Temporary.pathname })}
|
1147
|
+
|
1148
|
+
it "should save state in data.data subpath" do
|
1149
|
+
expect(plugin.last_run_metadata_file_path).to start_with(settings["last_run_metadata_path"])
|
1150
|
+
end
|
1151
|
+
end
|
1152
|
+
end
|
1153
|
+
|
1154
|
+
context "with customized last_run_metadata_path point to directory" do
|
1155
|
+
let(:settings) do
|
1156
|
+
path = Stud::Temporary.pathname
|
1157
|
+
Pathname.new(path).tap {|path| path.mkpath}
|
1158
|
+
super().merge({ "last_run_metadata_path" => path})
|
1159
|
+
end
|
1160
|
+
|
1161
|
+
it "raise configuration error" do
|
1162
|
+
expect { plugin.register }.to raise_error(LogStash::ConfigurationError)
|
1163
|
+
end
|
1164
|
+
end
|
1165
|
+
end
|
1166
|
+
|
1167
|
+
context "update the previous default last_run_metadata_path" do
|
1168
|
+
let(:settings) do
|
1169
|
+
{
|
1170
|
+
"statement" => "SELECT * FROM test_table",
|
1171
|
+
"record_last_run" => true
|
1172
|
+
}
|
1173
|
+
end
|
1174
|
+
|
1175
|
+
let(:fake_home) do
|
1176
|
+
path = Stud::Temporary.pathname
|
1177
|
+
Pathname.new(path).tap {|path| path.mkpath}
|
1178
|
+
path
|
1179
|
+
end
|
1180
|
+
|
1181
|
+
context "when a file exists" do
|
1182
|
+
before do
|
1183
|
+
# in a faked HOME folder save a valid previous last_run metadata file
|
1184
|
+
allow(ENV).to receive(:[]).with('HOME').and_return(fake_home)
|
1185
|
+
File.open("#{ENV['HOME']}/.logstash_jdbc_last_run", 'w') do |file|
|
1186
|
+
file.write("--- !ruby/object:DateTime '2022-03-08 08:10:00.486889000 Z'")
|
1187
|
+
end
|
1188
|
+
end
|
1189
|
+
|
1190
|
+
it "should be moved" do
|
1191
|
+
plugin.register
|
1192
|
+
|
1193
|
+
expect(::File.exist?("#{ENV['HOME']}/.logstash_jdbc_last_run")).to be false
|
1194
|
+
path = LogStash::SETTINGS.get_value("path.data")
|
1195
|
+
full_path = "#{path}/plugins/inputs/jdbc/logstash_jdbc_last_run"
|
1196
|
+
expect(::File.exist?(full_path)).to be true
|
1197
|
+
end
|
1198
|
+
end
|
1199
|
+
end
|
1200
|
+
|
1117
1201
|
context "when setting fetch size" do
|
1118
1202
|
|
1119
1203
|
let(:settings) do
|
@@ -1432,54 +1516,6 @@ describe LogStash::Inputs::Jdbc do
|
|
1432
1516
|
end
|
1433
1517
|
end
|
1434
1518
|
|
1435
|
-
context "when debug logging and a count query raises a count related error" do
|
1436
|
-
let(:settings) do
|
1437
|
-
{ "statement" => "SELECT * from types_table" }
|
1438
|
-
end
|
1439
|
-
let(:logger) { double("logger", :debug? => true) }
|
1440
|
-
let(:statement_logger) { LogStash::PluginMixins::Jdbc::CheckedCountLogger.new(logger) }
|
1441
|
-
let(:value_tracker) { double("value tracker", :set_value => nil, :write => nil) }
|
1442
|
-
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]' }
|
1443
|
-
let(:error_args) do
|
1444
|
-
{"exception" => msg}
|
1445
|
-
end
|
1446
|
-
|
1447
|
-
before do
|
1448
|
-
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)"
|
1449
|
-
plugin.register
|
1450
|
-
plugin.set_statement_logger(statement_logger)
|
1451
|
-
plugin.set_value_tracker(value_tracker)
|
1452
|
-
allow(value_tracker).to receive(:value).and_return("bar")
|
1453
|
-
allow(statement_logger).to receive(:execute_count).once.and_raise(StandardError.new(msg))
|
1454
|
-
end
|
1455
|
-
|
1456
|
-
after do
|
1457
|
-
plugin.stop
|
1458
|
-
end
|
1459
|
-
|
1460
|
-
context "if the count query raises an error" do
|
1461
|
-
it "should log a debug line without a count key as its unknown whether a count works at this stage" do
|
1462
|
-
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)
|
1463
|
-
expect(logger).to receive(:warn).once.with("Ongoing count statement generation is being prevented")
|
1464
|
-
expect(logger).to receive(:debug).once.with("Executing JDBC query", :statement => settings["statement"], :parameters => {:sql_last_value=>"bar"})
|
1465
|
-
plugin.run(queue)
|
1466
|
-
queue.pop
|
1467
|
-
end
|
1468
|
-
|
1469
|
-
it "should create an event normally" do
|
1470
|
-
allow(logger).to receive(:warn)
|
1471
|
-
allow(logger).to receive(:debug)
|
1472
|
-
plugin.run(queue)
|
1473
|
-
event = queue.pop
|
1474
|
-
expect(event.get("num")).to eq(1)
|
1475
|
-
expect(event.get("string")).to eq("A test")
|
1476
|
-
expect(event.get("started_at")).to be_a_logstash_timestamp_equivalent_to("1999-12-31T00:00:00.000Z")
|
1477
|
-
expect(event.get("custom_time")).to be_a_logstash_timestamp_equivalent_to("1999-12-31T23:59:59.000Z")
|
1478
|
-
expect(event.get("ranking").to_f).to eq(95.67)
|
1479
|
-
end
|
1480
|
-
end
|
1481
|
-
end
|
1482
|
-
|
1483
1519
|
context "when an unreadable jdbc_driver_path entry is present" do
|
1484
1520
|
let(:driver_jar_path) do
|
1485
1521
|
jar_file = $CLASSPATH.find { |name| name.index(Jdbc::Derby.driver_jar) }
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: logstash-integration-jdbc
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 5.2.
|
4
|
+
version: 5.2.6
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Elastic
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2022-
|
11
|
+
date: 2022-05-19 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
requirement: !ruby/object:Gem::Requirement
|
@@ -131,17 +131,17 @@ dependencies:
|
|
131
131
|
- !ruby/object:Gem::Dependency
|
132
132
|
requirement: !ruby/object:Gem::Requirement
|
133
133
|
requirements:
|
134
|
-
- - "
|
134
|
+
- - ">="
|
135
135
|
- !ruby/object:Gem::Version
|
136
|
-
version:
|
136
|
+
version: '0'
|
137
137
|
name: rufus-scheduler
|
138
138
|
prerelease: false
|
139
139
|
type: :runtime
|
140
140
|
version_requirements: !ruby/object:Gem::Requirement
|
141
141
|
requirements:
|
142
|
-
- - "
|
142
|
+
- - ">="
|
143
143
|
- !ruby/object:Gem::Version
|
144
|
-
version:
|
144
|
+
version: '0'
|
145
145
|
- !ruby/object:Gem::Dependency
|
146
146
|
requirement: !ruby/object:Gem::Requirement
|
147
147
|
requirements:
|
@@ -275,7 +275,6 @@ files:
|
|
275
275
|
- lib/logstash/filters/jdbc_streaming.rb
|
276
276
|
- lib/logstash/inputs/jdbc.rb
|
277
277
|
- lib/logstash/inputs/tzinfo_jruby_patch.rb
|
278
|
-
- lib/logstash/plugin_mixins/jdbc/checked_count_logger.rb
|
279
278
|
- lib/logstash/plugin_mixins/jdbc/common.rb
|
280
279
|
- lib/logstash/plugin_mixins/jdbc/jdbc.rb
|
281
280
|
- lib/logstash/plugin_mixins/jdbc/scheduler.rb
|
@@ -1,43 +0,0 @@
|
|
1
|
-
# encoding: utf-8
|
2
|
-
|
3
|
-
module LogStash module PluginMixins module Jdbc
|
4
|
-
class CheckedCountLogger
|
5
|
-
def initialize(logger)
|
6
|
-
@logger = logger
|
7
|
-
@needs_check = true
|
8
|
-
@count_is_supported = false
|
9
|
-
@in_debug = @logger.debug?
|
10
|
-
end
|
11
|
-
|
12
|
-
def disable_count
|
13
|
-
@needs_check = false
|
14
|
-
@count_is_supported = false
|
15
|
-
end
|
16
|
-
|
17
|
-
def log_statement_parameters(statement, parameters, query)
|
18
|
-
return unless @in_debug
|
19
|
-
check_count_query(query) if @needs_check && query
|
20
|
-
if @count_is_supported
|
21
|
-
@logger.debug("Executing JDBC query", :statement => statement, :parameters => parameters, :count => execute_count(query))
|
22
|
-
else
|
23
|
-
@logger.debug("Executing JDBC query", :statement => statement, :parameters => parameters)
|
24
|
-
end
|
25
|
-
end
|
26
|
-
|
27
|
-
def check_count_query(query)
|
28
|
-
@needs_check = false
|
29
|
-
begin
|
30
|
-
execute_count(query)
|
31
|
-
@count_is_supported = true
|
32
|
-
rescue Exception => e
|
33
|
-
@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)
|
34
|
-
@logger.warn("Ongoing count statement generation is being prevented")
|
35
|
-
@count_is_supported = false
|
36
|
-
end
|
37
|
-
end
|
38
|
-
|
39
|
-
def execute_count(query)
|
40
|
-
query.count
|
41
|
-
end
|
42
|
-
end
|
43
|
-
end end end
|