logstash-integration-jdbc 5.1.10 → 5.2.3
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG.md +15 -0
- data/docs/input-jdbc.asciidoc +51 -1
- data/lib/logstash/inputs/jdbc.rb +12 -2
- data/lib/logstash/plugin_mixins/jdbc/jdbc.rb +6 -2
- data/lib/logstash/plugin_mixins/jdbc/scheduler.rb +147 -0
- data/lib/logstash/plugin_mixins/jdbc/statement_handler.rb +61 -15
- data/logstash-integration-jdbc.gemspec +2 -3
- data/spec/inputs/integration/integ_spec.rb +1 -1
- data/spec/inputs/jdbc_spec.rb +33 -0
- data/spec/plugin_mixins/jdbc/scheduler_spec.rb +78 -0
- metadata +9 -6
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 2119dbab43322bdc083cf11602e69e63bcb1a957b537f5f44cfcd2b611df1893
|
4
|
+
data.tar.gz: 651e016e43bb17db730cc95ad0fd44bed483b9edfa4bbe86e52ced95947bb4a4
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 0a90d6aa88365c9ddfe60bba20f6f06633733af4b910b75becc31c37a9e1a224d3aae8d5e62f03f8ae2a3ffd8a24d1b0635307aa152a836c18435a4b36ca1d5d
|
7
|
+
data.tar.gz: f6a6648d071bff9069f54448cfed03327cb0f5bb8c036d8b0c6e5adb949c69a75f923541fe6a1237eeb3e6fe35611ad0afd4f99e356e8dd42b1e2b19534bc5b7
|
data/CHANGELOG.md
CHANGED
@@ -1,7 +1,22 @@
|
|
1
|
+
## 5.2.3
|
2
|
+
- Performance: avoid contention on scheduler execution [#103](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/103)
|
3
|
+
|
4
|
+
## 5.2.2
|
5
|
+
- Feat: name scheduler threads + redirect error logging [#102](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/102)
|
6
|
+
|
7
|
+
## 5.2.1
|
8
|
+
- Refactor: isolate paginated normal statement algorithm in a separate handler [#101](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/101)
|
9
|
+
|
10
|
+
## 5.2.0
|
11
|
+
- Added `jdbc_paging_mode` option to choose if use `explicit` pagination in statements and avoid the initial count
|
12
|
+
query or use `auto` to delegate to the underlying library [#95](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/95)
|
13
|
+
|
1
14
|
## 5.1.10
|
2
15
|
- Refactor: to explicit Java (driver) class name loading [#96](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/96),
|
3
16
|
the change is expected to provide a more robust fix for the driver loading issue [#83](https://github.com/logstash-plugins/logstash-integration-jdbc/issues/83).
|
4
17
|
|
18
|
+
NOTE: a fatal driver error will no longer keep reloading the pipeline and now leads to a system exit.
|
19
|
+
|
5
20
|
- Fix: regression due returning the Java driver class [#98](https://github.com/logstash-plugins/logstash-integration-jdbc/pull/98)
|
6
21
|
|
7
22
|
## 5.1.9 (yanked)
|
data/docs/input-jdbc.asciidoc
CHANGED
@@ -129,6 +129,9 @@ Here is the list:
|
|
129
129
|
|sql_last_value | The value used to calculate which rows to query. Before any query is run,
|
130
130
|
this is set to Thursday, 1 January 1970, or 0 if `use_column_value` is true and
|
131
131
|
`tracking_column` is set. It is updated accordingly after subsequent queries are run.
|
132
|
+
|offset, size| Values used with manual paging mode to explicitly implement the paging.
|
133
|
+
Supported only if <<plugins-{type}s-{plugin}-jdbc_paging_enabled>> is enabled and
|
134
|
+
<<plugins-{type}s-{plugin}-jdbc_paging_mode>> has the `explicit` value.
|
132
135
|
|==========================================================
|
133
136
|
|
134
137
|
Example:
|
@@ -153,7 +156,7 @@ NOTE: Not all JDBC accessible technologies will support prepared statements.
|
|
153
156
|
With the introduction of Prepared Statement support comes a different code execution path and some new settings. Most of the existing settings are still useful but there are several new settings for Prepared Statements to read up on.
|
154
157
|
Use the boolean setting `use_prepared_statements` to enable this execution mode. Use the `prepared_statement_name` setting to specify a name for the Prepared Statement, this identifies the prepared statement locally and remotely and it should be unique in your config and on the database. Use the `prepared_statement_bind_values` array setting to specify the bind values, use the exact string `:sql_last_value` (multiple times if necessary) for the predefined parameter mentioned before. The `statement` (or `statement_path`) setting still holds the SQL statement but to use bind variables you must use the `?` character as a placeholder in the exact order found in the `prepared_statement_bind_values` array.
|
155
158
|
|
156
|
-
NOTE: Building count queries around a prepared statement is not supported at this time
|
159
|
+
NOTE: Building count queries around a prepared statement is not supported at this time. Because jdbc paging uses count queries when `jdbc_paging_mode` has value `auto`,jdbc paging is not supported with prepared statements at this time either. Therefore, `jdbc_paging_enabled`, `jdbc_page_size` settings are ignored when using prepared statements.
|
157
160
|
|
158
161
|
Example:
|
159
162
|
[source,ruby]
|
@@ -193,6 +196,7 @@ This plugin supports the following configuration options plus the <<plugins-{typ
|
|
193
196
|
| <<plugins-{type}s-{plugin}-jdbc_fetch_size>> |<<number,number>>|No
|
194
197
|
| <<plugins-{type}s-{plugin}-jdbc_page_size>> |<<number,number>>|No
|
195
198
|
| <<plugins-{type}s-{plugin}-jdbc_paging_enabled>> |<<boolean,boolean>>|No
|
199
|
+
| <<plugins-{type}s-{plugin}-jdbc_paging_mode>> |<<string,string>>, one of `["auto", "explicit"]`|No
|
196
200
|
| <<plugins-{type}s-{plugin}-jdbc_password>> |<<password,password>>|No
|
197
201
|
| <<plugins-{type}s-{plugin}-jdbc_password_filepath>> |a valid filesystem path|No
|
198
202
|
| <<plugins-{type}s-{plugin}-jdbc_pool_timeout>> |<<number,number>>|No
|
@@ -373,6 +377,52 @@ result-set. The limit size is set with `jdbc_page_size`.
|
|
373
377
|
|
374
378
|
Be aware that ordering is not guaranteed between queries.
|
375
379
|
|
380
|
+
[id="plugins-{type}s-{plugin}-jdbc_paging_mode"]
|
381
|
+
===== `jdbc_paging_mode`
|
382
|
+
|
383
|
+
* Value can be any of: `auto`, `explicit`
|
384
|
+
* Default value is `"auto"`
|
385
|
+
|
386
|
+
Whether to use `explicit` or `auto` mode during the JDBC paging
|
387
|
+
|
388
|
+
If `auto`, your statement will be automatically surrounded by a count query and subsequent multiple paged queries (with `LIMIT` statement, etc.).
|
389
|
+
|
390
|
+
If `explicit`, multiple queries (without a count query ahead) will be performed with your statement, until no more rows are retrieved.
|
391
|
+
You have to write your own paging conditions in your statement configuration.
|
392
|
+
The `offset` and `size` parameters can be used in your statement (`size` equal to `jdbc_page_size`, and `offset` incremented by `size` for each query).
|
393
|
+
When the number of rows returned by the query is not equal to `size`, SQL paging will be ended.
|
394
|
+
Example:
|
395
|
+
|
396
|
+
[source, ruby]
|
397
|
+
------------------------------------------------------
|
398
|
+
input {
|
399
|
+
jdbc {
|
400
|
+
statement => "SELECT id, mycolumn1, mycolumn2 FROM my_table WHERE id > :sql_last_value LIMIT :size OFFSET :offset",
|
401
|
+
jdbc_paging_enabled => true,
|
402
|
+
jdbc_paging_mode => "explicit",
|
403
|
+
jdbc_page_size => 100000
|
404
|
+
}
|
405
|
+
}
|
406
|
+
------------------------------------------------------
|
407
|
+
|
408
|
+
[source, ruby]
|
409
|
+
------------------------------------------------------
|
410
|
+
input {
|
411
|
+
jdbc {
|
412
|
+
statement => "CALL fetch_my_data(:sql_last_value, :offset, :size)",
|
413
|
+
jdbc_paging_enabled => true,
|
414
|
+
jdbc_paging_mode => "explicit",
|
415
|
+
jdbc_page_size => 100000
|
416
|
+
}
|
417
|
+
}
|
418
|
+
------------------------------------------------------
|
419
|
+
|
420
|
+
This mode can be considered in the following situations:
|
421
|
+
|
422
|
+
. Performance issues encountered in default paging mode.
|
423
|
+
. Your SQL statement is complex, so simply surrounding it with paging statements is not what you want.
|
424
|
+
. Your statement is a stored procedure, and the actual paging statement is inside it.
|
425
|
+
|
376
426
|
[id="plugins-{type}s-{plugin}-jdbc_password"]
|
377
427
|
===== `jdbc_password`
|
378
428
|
|
data/lib/logstash/inputs/jdbc.rb
CHANGED
@@ -3,6 +3,7 @@ require "logstash/inputs/base"
|
|
3
3
|
require "logstash/namespace"
|
4
4
|
require "logstash/plugin_mixins/jdbc/common"
|
5
5
|
require "logstash/plugin_mixins/jdbc/jdbc"
|
6
|
+
require "logstash/plugin_mixins/jdbc/scheduler"
|
6
7
|
require "logstash/plugin_mixins/ecs_compatibility_support"
|
7
8
|
require "logstash/plugin_mixins/ecs_compatibility_support/target_check"
|
8
9
|
require "logstash/plugin_mixins/validator_support/field_reference_validation_adapter"
|
@@ -293,8 +294,17 @@ module LogStash module Inputs class Jdbc < LogStash::Inputs::Base
|
|
293
294
|
def run(queue)
|
294
295
|
load_driver
|
295
296
|
if @schedule
|
296
|
-
|
297
|
-
@scheduler
|
297
|
+
# input thread (Java) name example "[my-oracle]<jdbc"
|
298
|
+
@scheduler = LogStash::PluginMixins::Jdbc::Scheduler.new(
|
299
|
+
:max_work_threads => 1,
|
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
|
298
308
|
execute_query(queue)
|
299
309
|
end
|
300
310
|
|
@@ -55,6 +55,9 @@ module LogStash module PluginMixins module Jdbc
|
|
55
55
|
# Be aware that ordering is not guaranteed between queries.
|
56
56
|
config :jdbc_paging_enabled, :validate => :boolean, :default => false
|
57
57
|
|
58
|
+
# Which pagination mode to use, automatic pagination or explicitly defined in the query.
|
59
|
+
config :jdbc_paging_mode, :validate => [ "auto", "explicit" ], :default => "auto"
|
60
|
+
|
58
61
|
# JDBC page size
|
59
62
|
config :jdbc_page_size, :validate => :number, :default => 100000
|
60
63
|
|
@@ -211,13 +214,14 @@ module LogStash module PluginMixins module Jdbc
|
|
211
214
|
open_jdbc_connection
|
212
215
|
sql_last_value = @use_column_value ? @value_tracker.value : Time.now.utc
|
213
216
|
@tracking_column_warning_sent = false
|
214
|
-
@statement_handler.perform_query(@database, @value_tracker.value
|
217
|
+
@statement_handler.perform_query(@database, @value_tracker.value) do |row|
|
215
218
|
sql_last_value = get_column_value(row) if @use_column_value
|
216
219
|
yield extract_values_from(row)
|
217
220
|
end
|
218
221
|
success = true
|
219
222
|
rescue Sequel::DatabaseConnectionError, Sequel::DatabaseError, Java::JavaSql::SQLException => e
|
220
|
-
details = { :
|
223
|
+
details = { exception: e.class, message: e.message }
|
224
|
+
details[:cause] = e.cause.inspect if e.cause
|
221
225
|
details[:backtrace] = e.backtrace if @logger.debug?
|
222
226
|
@logger.warn("Exception when executing JDBC query", details)
|
223
227
|
else
|
@@ -0,0 +1,147 @@
|
|
1
|
+
require 'rufus/scheduler'
|
2
|
+
|
3
|
+
require 'logstash/util/loggable'
|
4
|
+
|
5
|
+
module LogStash module PluginMixins module Jdbc
|
6
|
+
class Scheduler < Rufus::Scheduler
|
7
|
+
|
8
|
+
include LogStash::Util::Loggable
|
9
|
+
|
10
|
+
# Rufus::Scheduler >= 3.4 moved the Time impl into a gem EoTime = ::EtOrbi::EoTime`
|
11
|
+
# Rufus::Scheduler 3.1 - 3.3 using it's own Time impl `Rufus::Scheduler::ZoTime`
|
12
|
+
TimeImpl = defined?(Rufus::Scheduler::EoTime) ? Rufus::Scheduler::EoTime :
|
13
|
+
(defined?(Rufus::Scheduler::ZoTime) ? Rufus::Scheduler::ZoTime : ::Time)
|
14
|
+
|
15
|
+
# @overload
|
16
|
+
def timeout_jobs
|
17
|
+
# Rufus relies on `Thread.list` which is a blocking operation and with many schedulers
|
18
|
+
# (and threads) within LS will have a negative impact on performance as scheduler
|
19
|
+
# threads will end up waiting to obtain the `Thread.list` lock.
|
20
|
+
#
|
21
|
+
# However, this isn't necessary we can easily detect whether there are any jobs
|
22
|
+
# that might need to timeout: only when `@opts[:timeout]` is set causes worker thread(s)
|
23
|
+
# to have a `Thread.current[:rufus_scheduler_timeout]` that is not nil
|
24
|
+
return unless @opts[:timeout]
|
25
|
+
super
|
26
|
+
end
|
27
|
+
|
28
|
+
# @overload
|
29
|
+
def work_threads(query = :all)
|
30
|
+
if query == :__all_no_cache__ # special case from JobDecorator#start_work_thread
|
31
|
+
@_work_threads = nil # when a new worker thread is being added reset
|
32
|
+
return super(:all)
|
33
|
+
end
|
34
|
+
|
35
|
+
# Gets executed every time a job is triggered, we're going to cache the
|
36
|
+
# worker threads for this scheduler (to avoid `Thread.list`) - they only
|
37
|
+
# change when a new thread is being started from #start_work_thread ...
|
38
|
+
work_threads = @_work_threads
|
39
|
+
if work_threads.nil?
|
40
|
+
work_threads = threads.select { |t| t[:rufus_scheduler_work_thread] }
|
41
|
+
@_work_threads = work_threads
|
42
|
+
end
|
43
|
+
|
44
|
+
case query
|
45
|
+
when :active then work_threads.select { |t| t[:rufus_scheduler_job] }
|
46
|
+
when :vacant then work_threads.reject { |t| t[:rufus_scheduler_job] }
|
47
|
+
else work_threads
|
48
|
+
end
|
49
|
+
end
|
50
|
+
|
51
|
+
# @overload
|
52
|
+
def on_error(job, err)
|
53
|
+
details = { exception: err.class, message: err.message, backtrace: err.backtrace }
|
54
|
+
details[:cause] = err.cause if err.cause
|
55
|
+
|
56
|
+
details[:now] = debug_format_time(TimeImpl.now)
|
57
|
+
details[:last_time] = (debug_format_time(job.last_time) rescue nil)
|
58
|
+
details[:next_time] = (debug_format_time(job.next_time) rescue nil)
|
59
|
+
details[:job] = job
|
60
|
+
|
61
|
+
details[:opts] = @opts
|
62
|
+
details[:started_at] = started_at
|
63
|
+
details[:thread] = thread.inspect
|
64
|
+
details[:jobs_size] = @jobs.size
|
65
|
+
details[:work_threads_size] = work_threads.size
|
66
|
+
details[:work_queue_size] = work_queue.size
|
67
|
+
|
68
|
+
logger.error("Scheduler intercepted an error:", details)
|
69
|
+
|
70
|
+
rescue => e
|
71
|
+
logger.error("Scheduler failed in #on_error #{e.inspect}")
|
72
|
+
end
|
73
|
+
|
74
|
+
def debug_format_time(time)
|
75
|
+
# EtOrbi::EoTime used by (newer) Rufus::Scheduler has to_debug_s https://git.io/JyiPj
|
76
|
+
time.respond_to?(:to_debug_s) ? time.to_debug_s : time.strftime("%Y-%m-%dT%H:%M:%S.%L")
|
77
|
+
end
|
78
|
+
private :debug_format_time
|
79
|
+
|
80
|
+
# @private helper used by JobDecorator
|
81
|
+
def work_thread_name_prefix
|
82
|
+
( @opts[:thread_name] || "#{@thread_key}_scheduler" ) + '_worker-'
|
83
|
+
end
|
84
|
+
|
85
|
+
protected
|
86
|
+
|
87
|
+
# @overload
|
88
|
+
def start
|
89
|
+
ret = super() # @thread[:name] = @opts[:thread_name] || "#{@thread_key}_scheduler"
|
90
|
+
|
91
|
+
# at least set thread.name for easier thread dump analysis
|
92
|
+
if @thread.is_a?(Thread) && @thread.respond_to?(:name=)
|
93
|
+
@thread.name = @thread[:name] if @thread[:name]
|
94
|
+
end
|
95
|
+
|
96
|
+
ret
|
97
|
+
end
|
98
|
+
|
99
|
+
# @overload
|
100
|
+
def do_schedule(job_type, t, callable, opts, return_job_instance, block)
|
101
|
+
job_or_id = super
|
102
|
+
|
103
|
+
job_or_id.extend JobDecorator if return_job_instance
|
104
|
+
|
105
|
+
job_or_id
|
106
|
+
end
|
107
|
+
|
108
|
+
module JobDecorator
|
109
|
+
|
110
|
+
def start_work_thread
|
111
|
+
prev_thread_count = @scheduler.work_threads.size
|
112
|
+
|
113
|
+
ret = super() # does not return Thread instance in 3.0
|
114
|
+
|
115
|
+
work_threads = @scheduler.work_threads(:__all_no_cache__)
|
116
|
+
while prev_thread_count == work_threads.size # very unlikely
|
117
|
+
Thread.pass
|
118
|
+
work_threads = @scheduler.work_threads(:__all_no_cache__)
|
119
|
+
end
|
120
|
+
|
121
|
+
work_thread_name_prefix = @scheduler.work_thread_name_prefix
|
122
|
+
|
123
|
+
work_threads.sort! do |t1, t2|
|
124
|
+
if t1[:name].nil?
|
125
|
+
t2[:name].nil? ? 0 : +1 # nils at the end
|
126
|
+
elsif t2[:name].nil?
|
127
|
+
t1[:name].nil? ? 0 : -1
|
128
|
+
else
|
129
|
+
t1[:name] <=> t2[:name]
|
130
|
+
end
|
131
|
+
end
|
132
|
+
|
133
|
+
work_threads.each_with_index do |thread, i|
|
134
|
+
unless thread[:name]
|
135
|
+
thread[:name] = "#{work_thread_name_prefix}#{sprintf('%02i', i)}"
|
136
|
+
thread.name = thread[:name] if thread.respond_to?(:name=)
|
137
|
+
# e.g. "[oracle]<jdbc_scheduler_worker-00"
|
138
|
+
end
|
139
|
+
end
|
140
|
+
|
141
|
+
ret
|
142
|
+
end
|
143
|
+
|
144
|
+
end
|
145
|
+
|
146
|
+
end
|
147
|
+
end end end
|
@@ -3,7 +3,19 @@
|
|
3
3
|
module LogStash module PluginMixins module Jdbc
|
4
4
|
class StatementHandler
|
5
5
|
def self.build_statement_handler(plugin, logger)
|
6
|
-
|
6
|
+
if plugin.use_prepared_statements
|
7
|
+
klass = PreparedStatementHandler
|
8
|
+
else
|
9
|
+
if plugin.jdbc_paging_enabled
|
10
|
+
if plugin.jdbc_paging_mode == "explicit"
|
11
|
+
klass = ExplicitPagingModeStatementHandler
|
12
|
+
else
|
13
|
+
klass = PagedNormalStatementHandler
|
14
|
+
end
|
15
|
+
else
|
16
|
+
klass = NormalStatementHandler
|
17
|
+
end
|
18
|
+
end
|
7
19
|
klass.new(plugin, logger)
|
8
20
|
end
|
9
21
|
|
@@ -25,22 +37,14 @@ module LogStash module PluginMixins module Jdbc
|
|
25
37
|
end
|
26
38
|
|
27
39
|
class NormalStatementHandler < StatementHandler
|
28
|
-
# Performs the query,
|
40
|
+
# Performs the query, yielding once per row of data
|
29
41
|
# @param db [Sequel::Database]
|
30
|
-
# @param sql_last_value [
|
42
|
+
# @param sql_last_value [Integer|DateTime|Time]
|
31
43
|
# @yieldparam row [Hash{Symbol=>Object}]
|
32
|
-
def perform_query(db, sql_last_value
|
44
|
+
def perform_query(db, sql_last_value)
|
33
45
|
query = build_query(db, sql_last_value)
|
34
|
-
|
35
|
-
|
36
|
-
paged_dataset.each do |row|
|
37
|
-
yield row
|
38
|
-
end
|
39
|
-
end
|
40
|
-
else
|
41
|
-
query.each do |row|
|
42
|
-
yield row
|
43
|
-
end
|
46
|
+
query.each do |row|
|
47
|
+
yield row
|
44
48
|
end
|
45
49
|
end
|
46
50
|
|
@@ -67,6 +71,48 @@ module LogStash module PluginMixins module Jdbc
|
|
67
71
|
end
|
68
72
|
end
|
69
73
|
|
74
|
+
class PagedNormalStatementHandler < NormalStatementHandler
|
75
|
+
attr_reader :jdbc_page_size
|
76
|
+
|
77
|
+
# Performs the query, respecting our pagination settings, yielding once per row of data
|
78
|
+
# @param db [Sequel::Database]
|
79
|
+
# @param sql_last_value [Integer|DateTime|Time]
|
80
|
+
# @yieldparam row [Hash{Symbol=>Object}]
|
81
|
+
def perform_query(db, sql_last_value)
|
82
|
+
query = build_query(db, sql_last_value)
|
83
|
+
query.each_page(@jdbc_page_size) do |paged_dataset|
|
84
|
+
paged_dataset.each do |row|
|
85
|
+
yield row
|
86
|
+
end
|
87
|
+
end
|
88
|
+
end
|
89
|
+
|
90
|
+
def post_init(plugin)
|
91
|
+
super(plugin)
|
92
|
+
@jdbc_page_size = plugin.jdbc_page_size
|
93
|
+
end
|
94
|
+
end
|
95
|
+
|
96
|
+
class ExplicitPagingModeStatementHandler < PagedNormalStatementHandler
|
97
|
+
# Performs the query, respecting our pagination settings, yielding once per row of data
|
98
|
+
# @param db [Sequel::Database]
|
99
|
+
# @param sql_last_value [Integer|DateTime|Time]
|
100
|
+
# @yieldparam row [Hash{Symbol=>Object}]
|
101
|
+
def perform_query(db, sql_last_value)
|
102
|
+
query = build_query(db, sql_last_value)
|
103
|
+
offset = 0
|
104
|
+
loop do
|
105
|
+
rows_in_page = 0
|
106
|
+
query.with_sql(query.sql, offset: offset, size: jdbc_page_size).each do |row|
|
107
|
+
yield row
|
108
|
+
rows_in_page += 1
|
109
|
+
end
|
110
|
+
break unless rows_in_page == jdbc_page_size
|
111
|
+
offset += jdbc_page_size
|
112
|
+
end
|
113
|
+
end
|
114
|
+
end
|
115
|
+
|
70
116
|
class PreparedStatementHandler < StatementHandler
|
71
117
|
attr_reader :name, :bind_values_array, :statement_prepared, :prepared
|
72
118
|
|
@@ -74,7 +120,7 @@ module LogStash module PluginMixins module Jdbc
|
|
74
120
|
# @param db [Sequel::Database]
|
75
121
|
# @param sql_last_value [Integet|DateTime|Time]
|
76
122
|
# @yieldparam row [Hash{Symbol=>Object}]
|
77
|
-
def perform_query(db, sql_last_value
|
123
|
+
def perform_query(db, sql_last_value)
|
78
124
|
query = build_query(db, sql_last_value)
|
79
125
|
query.each do |row|
|
80
126
|
yield row
|
@@ -1,6 +1,6 @@
|
|
1
1
|
Gem::Specification.new do |s|
|
2
2
|
s.name = 'logstash-integration-jdbc'
|
3
|
-
s.version = '5.
|
3
|
+
s.version = '5.2.3'
|
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"
|
@@ -34,8 +34,7 @@ Gem::Specification.new do |s|
|
|
34
34
|
|
35
35
|
s.add_runtime_dependency 'tzinfo'
|
36
36
|
s.add_runtime_dependency 'tzinfo-data'
|
37
|
-
|
38
|
-
s.add_runtime_dependency 'rufus-scheduler', '< 3.5'
|
37
|
+
s.add_runtime_dependency 'rufus-scheduler', '~> 3.0.9'
|
39
38
|
s.add_runtime_dependency 'logstash-mixin-ecs_compatibility_support', '~>1.3'
|
40
39
|
s.add_runtime_dependency "logstash-mixin-validator_support", '~> 1.0'
|
41
40
|
s.add_runtime_dependency "logstash-mixin-event_support", '~> 1.0'
|
@@ -70,7 +70,7 @@ describe LogStash::Inputs::Jdbc, :integration => true do
|
|
70
70
|
plugin.register
|
71
71
|
expect( plugin ).to receive(:log_java_exception)
|
72
72
|
expect(plugin.logger).to receive(:warn).once.with("Exception when executing JDBC query",
|
73
|
-
hash_including(:
|
73
|
+
hash_including(:message => instance_of(String)))
|
74
74
|
q = Queue.new
|
75
75
|
expect{ plugin.run(q) }.not_to raise_error
|
76
76
|
end
|
data/spec/inputs/jdbc_spec.rb
CHANGED
@@ -329,6 +329,39 @@ describe LogStash::Inputs::Jdbc do
|
|
329
329
|
|
330
330
|
end
|
331
331
|
|
332
|
+
context "when iterating result-set via explicit paging mode" do
|
333
|
+
|
334
|
+
let(:settings) do
|
335
|
+
{
|
336
|
+
"statement" => "SELECT * from test_table OFFSET :offset ROWS FETCH NEXT :size ROWS ONLY",
|
337
|
+
"jdbc_paging_enabled" => true,
|
338
|
+
"jdbc_paging_mode" => "explicit",
|
339
|
+
"jdbc_page_size" => 10
|
340
|
+
}
|
341
|
+
end
|
342
|
+
|
343
|
+
let(:num_rows) { 15 }
|
344
|
+
|
345
|
+
before do
|
346
|
+
plugin.register
|
347
|
+
end
|
348
|
+
|
349
|
+
after do
|
350
|
+
plugin.stop
|
351
|
+
end
|
352
|
+
|
353
|
+
it "should fetch all rows" do
|
354
|
+
num_rows.times do
|
355
|
+
db[:test_table].insert(:num => 1, :custom_time => Time.now.utc, :created_at => Time.now.utc)
|
356
|
+
end
|
357
|
+
|
358
|
+
plugin.run(queue)
|
359
|
+
|
360
|
+
expect(queue.size).to eq(num_rows)
|
361
|
+
end
|
362
|
+
|
363
|
+
end
|
364
|
+
|
332
365
|
context "when using target option" do
|
333
366
|
let(:settings) do
|
334
367
|
{
|
@@ -0,0 +1,78 @@
|
|
1
|
+
# encoding: utf-8
|
2
|
+
require "logstash/devutils/rspec/spec_helper"
|
3
|
+
require "logstash/plugin_mixins/jdbc/scheduler"
|
4
|
+
|
5
|
+
describe LogStash::PluginMixins::Jdbc::Scheduler do
|
6
|
+
|
7
|
+
let(:thread_name) { '[test]<jdbc_scheduler' }
|
8
|
+
|
9
|
+
let(:opts) do
|
10
|
+
{ :max_work_threads => 2, :thread_name => thread_name }
|
11
|
+
end
|
12
|
+
|
13
|
+
subject(:scheduler) { LogStash::PluginMixins::Jdbc::Scheduler.new(opts) }
|
14
|
+
|
15
|
+
after { scheduler.stop(:wait) }
|
16
|
+
|
17
|
+
it "sets scheduler thread name" do
|
18
|
+
expect( scheduler.thread.name ).to include thread_name
|
19
|
+
end
|
20
|
+
|
21
|
+
context 'cron schedule' do
|
22
|
+
|
23
|
+
before do
|
24
|
+
scheduler.schedule_cron('* * * * * *') { sleep 1.25 } # every second
|
25
|
+
end
|
26
|
+
|
27
|
+
it "sets worker thread names" do
|
28
|
+
sleep 3.0
|
29
|
+
threads = scheduler.work_threads
|
30
|
+
threads.sort! { |t1, t2| (t1.name || '') <=> (t2.name || '') }
|
31
|
+
|
32
|
+
expect( threads.size ).to eql 2
|
33
|
+
expect( threads.first.name ).to eql "#{thread_name}_worker-00"
|
34
|
+
expect( threads.last.name ).to eql "#{thread_name}_worker-01"
|
35
|
+
end
|
36
|
+
|
37
|
+
end
|
38
|
+
|
39
|
+
context 'every 1s' do
|
40
|
+
|
41
|
+
before do
|
42
|
+
scheduler.schedule_in('1s') { raise 'TEST' } # every second
|
43
|
+
end
|
44
|
+
|
45
|
+
it "logs errors handled" do
|
46
|
+
expect( scheduler.logger ).to receive(:error).with /Scheduler intercepted an error/, hash_including(:message => 'TEST')
|
47
|
+
sleep 1.5
|
48
|
+
end
|
49
|
+
|
50
|
+
end
|
51
|
+
|
52
|
+
context 'work threads' do
|
53
|
+
|
54
|
+
let(:opts) { super().merge :max_work_threads => 3 }
|
55
|
+
|
56
|
+
let(:counter) { java.util.concurrent.atomic.AtomicLong.new(0) }
|
57
|
+
|
58
|
+
before do
|
59
|
+
scheduler.schedule_cron('* * * * * *') { counter.increment_and_get; sleep 3.25 } # every second
|
60
|
+
end
|
61
|
+
|
62
|
+
it "are working" do
|
63
|
+
sleep(0.05) while counter.get == 0
|
64
|
+
expect( scheduler.work_threads.size ).to eql 1
|
65
|
+
sleep(0.05) while counter.get == 1
|
66
|
+
expect( scheduler.work_threads.size ).to eql 2
|
67
|
+
sleep(0.05) while counter.get == 2
|
68
|
+
expect( scheduler.work_threads.size ).to eql 3
|
69
|
+
|
70
|
+
sleep 1.25
|
71
|
+
expect( scheduler.work_threads.size ).to eql 3
|
72
|
+
sleep 1.25
|
73
|
+
expect( scheduler.work_threads.size ).to eql 3
|
74
|
+
end
|
75
|
+
|
76
|
+
end
|
77
|
+
|
78
|
+
end
|
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.
|
4
|
+
version: 5.2.3
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Elastic
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date:
|
11
|
+
date: 2022-02-16 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: 3.0.9
|
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: 3.0.9
|
145
145
|
- !ruby/object:Gem::Dependency
|
146
146
|
requirement: !ruby/object:Gem::Requirement
|
147
147
|
requirements:
|
@@ -278,6 +278,7 @@ files:
|
|
278
278
|
- lib/logstash/plugin_mixins/jdbc/checked_count_logger.rb
|
279
279
|
- lib/logstash/plugin_mixins/jdbc/common.rb
|
280
280
|
- lib/logstash/plugin_mixins/jdbc/jdbc.rb
|
281
|
+
- lib/logstash/plugin_mixins/jdbc/scheduler.rb
|
281
282
|
- lib/logstash/plugin_mixins/jdbc/statement_handler.rb
|
282
283
|
- lib/logstash/plugin_mixins/jdbc/value_tracking.rb
|
283
284
|
- lib/logstash/plugin_mixins/jdbc_streaming.rb
|
@@ -306,6 +307,7 @@ files:
|
|
306
307
|
- spec/helpers/derbyrun.jar
|
307
308
|
- spec/inputs/integration/integ_spec.rb
|
308
309
|
- spec/inputs/jdbc_spec.rb
|
310
|
+
- spec/plugin_mixins/jdbc/scheduler_spec.rb
|
309
311
|
- spec/plugin_mixins/jdbc_streaming/parameter_handler_spec.rb
|
310
312
|
- vendor/jar-dependencies/org/apache/derby/derby/10.14.1.0/derby-10.14.1.0.jar
|
311
313
|
- vendor/jar-dependencies/org/apache/derby/derbyclient/10.14.1.0/derbyclient-10.14.1.0.jar
|
@@ -358,4 +360,5 @@ test_files:
|
|
358
360
|
- spec/helpers/derbyrun.jar
|
359
361
|
- spec/inputs/integration/integ_spec.rb
|
360
362
|
- spec/inputs/jdbc_spec.rb
|
363
|
+
- spec/plugin_mixins/jdbc/scheduler_spec.rb
|
361
364
|
- spec/plugin_mixins/jdbc_streaming/parameter_handler_spec.rb
|