karafka 2.0.0.beta4 → 2.0.0.rc2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- checksums.yaml.gz.sig +0 -0
- data/.github/workflows/ci.yml +18 -1
- data/CHANGELOG.md +30 -0
- data/CONTRIBUTING.md +0 -5
- data/Gemfile.lock +12 -42
- data/README.md +2 -12
- data/bin/benchmarks +2 -2
- data/bin/integrations +10 -3
- data/bin/{stress → stress_many} +1 -1
- data/bin/stress_one +13 -0
- data/config/errors.yml +48 -5
- data/docker-compose.yml +27 -18
- data/karafka.gemspec +2 -4
- data/lib/karafka/active_job/job_options_contract.rb +8 -2
- data/lib/karafka/active_job/routing/extensions.rb +1 -1
- data/lib/karafka/app.rb +2 -1
- data/lib/karafka/base_consumer.rb +24 -19
- data/lib/karafka/cli/install.rb +15 -2
- data/lib/karafka/cli/server.rb +4 -2
- data/lib/karafka/connection/client.rb +40 -17
- data/lib/karafka/connection/listener.rb +37 -11
- data/lib/karafka/connection/rebalance_manager.rb +20 -19
- data/lib/karafka/contracts/base.rb +2 -8
- data/lib/karafka/contracts/config.rb +71 -38
- data/lib/karafka/contracts/consumer_group.rb +25 -18
- data/lib/karafka/contracts/consumer_group_topic.rb +30 -16
- data/lib/karafka/contracts/server_cli_options.rb +18 -7
- data/lib/karafka/errors.rb +3 -0
- data/lib/karafka/helpers/colorize.rb +20 -0
- data/lib/karafka/pro/active_job/consumer.rb +1 -8
- data/lib/karafka/pro/active_job/job_options_contract.rb +10 -6
- data/lib/karafka/pro/base_consumer.rb +27 -21
- data/lib/karafka/pro/loader.rb +13 -6
- data/lib/karafka/pro/processing/coordinator.rb +63 -0
- data/lib/karafka/pro/processing/jobs_builder.rb +3 -2
- data/lib/karafka/pro/processing/partitioner.rb +41 -0
- data/lib/karafka/pro/processing/scheduler.rb +56 -0
- data/lib/karafka/pro/routing/extensions.rb +6 -0
- data/lib/karafka/processing/coordinator.rb +88 -0
- data/lib/karafka/processing/coordinators_buffer.rb +54 -0
- data/lib/karafka/processing/executor.rb +7 -17
- data/lib/karafka/processing/executors_buffer.rb +46 -15
- data/lib/karafka/processing/jobs/consume.rb +4 -2
- data/lib/karafka/processing/jobs_builder.rb +3 -2
- data/lib/karafka/processing/partitioner.rb +22 -0
- data/lib/karafka/processing/result.rb +0 -5
- data/lib/karafka/processing/scheduler.rb +22 -0
- data/lib/karafka/routing/consumer_group.rb +1 -1
- data/lib/karafka/routing/topic.rb +9 -0
- data/lib/karafka/setup/config.rb +26 -12
- data/lib/karafka/templates/example_consumer.rb.erb +2 -2
- data/lib/karafka/version.rb +1 -1
- data/lib/karafka.rb +0 -2
- data.tar.gz.sig +0 -0
- metadata +15 -36
- metadata.gz.sig +0 -0
- data/lib/karafka/pro/scheduler.rb +0 -54
- data/lib/karafka/scheduler.rb +0 -20
@@ -12,17 +12,19 @@ module Karafka
|
|
12
12
|
# @param executor [Karafka::Processing::Executor] executor that is suppose to run a given
|
13
13
|
# job
|
14
14
|
# @param messages [Karafka::Messages::Messages] karafka messages batch
|
15
|
+
# @param coordinator [Karafka::Processing::Coordinator] processing coordinator
|
15
16
|
# @return [Consume]
|
16
|
-
def initialize(executor, messages)
|
17
|
+
def initialize(executor, messages, coordinator)
|
17
18
|
@executor = executor
|
18
19
|
@messages = messages
|
20
|
+
@coordinator = coordinator
|
19
21
|
@created_at = Time.now
|
20
22
|
super()
|
21
23
|
end
|
22
24
|
|
23
25
|
# Runs the before consumption preparations on the executor
|
24
26
|
def before_call
|
25
|
-
executor.before_consume(@messages, @created_at)
|
27
|
+
executor.before_consume(@messages, @created_at, @coordinator)
|
26
28
|
end
|
27
29
|
|
28
30
|
# Runs the given executor
|
@@ -7,9 +7,10 @@ module Karafka
|
|
7
7
|
class JobsBuilder
|
8
8
|
# @param executor [Karafka::Processing::Executor]
|
9
9
|
# @param messages [Karafka::Messages::Messages] messages batch to be consumed
|
10
|
+
# @param coordinator [Karafka::Processing::Coordinator]
|
10
11
|
# @return [Karafka::Processing::Jobs::Consume] consumption job
|
11
|
-
def consume(executor, messages)
|
12
|
-
Jobs::Consume.new(executor, messages)
|
12
|
+
def consume(executor, messages, coordinator)
|
13
|
+
Jobs::Consume.new(executor, messages, coordinator)
|
13
14
|
end
|
14
15
|
|
15
16
|
# @param executor [Karafka::Processing::Executor]
|
@@ -0,0 +1,22 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module Karafka
|
4
|
+
module Processing
|
5
|
+
# Basic partitioner for work division
|
6
|
+
# It does not divide any work.
|
7
|
+
class Partitioner
|
8
|
+
# @param subscription_group [Karafka::Routing::SubscriptionGroup] subscription group
|
9
|
+
def initialize(subscription_group)
|
10
|
+
@subscription_group = subscription_group
|
11
|
+
end
|
12
|
+
|
13
|
+
# @param _topic [String] topic name
|
14
|
+
# @param messages [Array<Karafka::Messages::Message>] karafka messages
|
15
|
+
# @yieldparam [Integer] group id
|
16
|
+
# @yieldparam [Array<Karafka::Messages::Message>] karafka messages
|
17
|
+
def call(_topic, messages)
|
18
|
+
yield(0, messages)
|
19
|
+
end
|
20
|
+
end
|
21
|
+
end
|
22
|
+
end
|
@@ -0,0 +1,22 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
module Karafka
|
4
|
+
module Processing
|
5
|
+
# FIFO scheduler for messages coming from various topics and partitions
|
6
|
+
class Scheduler
|
7
|
+
# Schedules jobs in the fifo order
|
8
|
+
#
|
9
|
+
# @param queue [Karafka::Processing::JobsQueue] queue where we want to put the jobs
|
10
|
+
# @param jobs_array [Array<Karafka::Processing::Jobs::Base>] jobs we want to schedule
|
11
|
+
def schedule_consumption(queue, jobs_array)
|
12
|
+
jobs_array.each do |job|
|
13
|
+
queue << job
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
# Both revocation and shutdown jobs can also run in fifo by default
|
18
|
+
alias schedule_revocation schedule_consumption
|
19
|
+
alias schedule_shutdown schedule_consumption
|
20
|
+
end
|
21
|
+
end
|
22
|
+
end
|
@@ -38,7 +38,7 @@ module Karafka
|
|
38
38
|
# @return [Array<Routing::SubscriptionGroup>] all the subscription groups build based on
|
39
39
|
# the consumer group topics
|
40
40
|
def subscription_groups
|
41
|
-
App.config.internal.subscription_groups_builder.call(topics)
|
41
|
+
App.config.internal.routing.subscription_groups_builder.call(topics)
|
42
42
|
end
|
43
43
|
|
44
44
|
# Hashed version of consumer group that can be used for validation purposes
|
@@ -66,6 +66,15 @@ module Karafka
|
|
66
66
|
end
|
67
67
|
end
|
68
68
|
|
69
|
+
# @return [Class] consumer class that we should use
|
70
|
+
# @note This is just an alias to the `#consumer` method. We however want to use it internally
|
71
|
+
# instead of referencing the `#consumer`. We use this to indicate that this method returns
|
72
|
+
# class and not an instance. In the routing we want to keep the `#consumer Consumer`
|
73
|
+
# routing syntax, but for references outside, we should use this one.
|
74
|
+
def consumer_class
|
75
|
+
consumer
|
76
|
+
end
|
77
|
+
|
69
78
|
# @return [Boolean] true if this topic offset is handled by the end user
|
70
79
|
def manual_offset_management?
|
71
80
|
manual_offset_management
|
data/lib/karafka/setup/config.rb
CHANGED
@@ -12,7 +12,7 @@ module Karafka
|
|
12
12
|
# enough and will still keep the code simple
|
13
13
|
# @see Karafka::Setup::Configurators::Base for more details about configurators api
|
14
14
|
class Config
|
15
|
-
extend
|
15
|
+
extend ::WaterDrop::Configurable
|
16
16
|
|
17
17
|
# Defaults for kafka settings, that will be overwritten only if not present already
|
18
18
|
KAFKA_DEFAULTS = {
|
@@ -62,7 +62,7 @@ module Karafka
|
|
62
62
|
# options max_messages [Integer] how many messages do we want to fetch from Kafka in one go
|
63
63
|
setting :max_messages, default: 1_000
|
64
64
|
# option [Integer] number of milliseconds we can wait while fetching data
|
65
|
-
setting :max_wait_time, default:
|
65
|
+
setting :max_wait_time, default: 5_000
|
66
66
|
# option shutdown_timeout [Integer] the number of milliseconds after which Karafka no
|
67
67
|
# longer waits for the consumers to stop gracefully but instead we force terminate
|
68
68
|
# everything.
|
@@ -85,21 +85,31 @@ module Karafka
|
|
85
85
|
|
86
86
|
# Namespace for internal settings that should not be modified directly
|
87
87
|
setting :internal do
|
88
|
-
# option routing_builder [Karafka::Routing::Builder] builder instance
|
89
|
-
setting :routing_builder, default: Routing::Builder.new
|
90
88
|
# option status [Karafka::Status] app status
|
91
89
|
setting :status, default: Status.new
|
92
90
|
# option process [Karafka::Process] process status
|
93
91
|
# @note In the future, we need to have a single process representation for all the karafka
|
94
92
|
# instances
|
95
93
|
setting :process, default: Process.new
|
96
|
-
|
97
|
-
|
98
|
-
|
99
|
-
|
100
|
-
|
101
|
-
|
102
|
-
|
94
|
+
|
95
|
+
setting :routing do
|
96
|
+
# option builder [Karafka::Routing::Builder] builder instance
|
97
|
+
setting :builder, default: Routing::Builder.new
|
98
|
+
# option subscription_groups_builder [Routing::SubscriptionGroupsBuilder] subscription
|
99
|
+
# group builder
|
100
|
+
setting :subscription_groups_builder, default: Routing::SubscriptionGroupsBuilder.new
|
101
|
+
end
|
102
|
+
|
103
|
+
setting :processing do
|
104
|
+
# option scheduler [Object] scheduler we will be using
|
105
|
+
setting :scheduler, default: Processing::Scheduler.new
|
106
|
+
# option jobs_builder [Object] jobs builder we want to use
|
107
|
+
setting :jobs_builder, default: Processing::JobsBuilder.new
|
108
|
+
# option coordinator [Class] work coordinator we want to user for processing coordination
|
109
|
+
setting :coordinator_class, default: Processing::Coordinator
|
110
|
+
# option partitioner_class [Class] partitioner we use against a batch of data
|
111
|
+
setting :partitioner_class, default: Processing::Partitioner
|
112
|
+
end
|
103
113
|
|
104
114
|
# Karafka components for ActiveJob
|
105
115
|
setting :active_job do
|
@@ -109,10 +119,14 @@ module Karafka
|
|
109
119
|
# ensuring, that extra job options defined are valid
|
110
120
|
setting :job_options_contract, default: ActiveJob::JobOptionsContract.new
|
111
121
|
# option consumer [Class] consumer class that should be used to consume ActiveJob data
|
112
|
-
setting :
|
122
|
+
setting :consumer_class, default: ActiveJob::Consumer
|
113
123
|
end
|
114
124
|
end
|
115
125
|
|
126
|
+
# This will load all the defaults that can be later overwritten.
|
127
|
+
# Thanks to that we have an initial state out of the box.
|
128
|
+
configure
|
129
|
+
|
116
130
|
class << self
|
117
131
|
# Configuring method
|
118
132
|
# @param block [Proc] block we want to execute with the config instance
|
@@ -7,10 +7,10 @@ class ExampleConsumer < ApplicationConsumer
|
|
7
7
|
end
|
8
8
|
|
9
9
|
# Run anything upon partition being revoked
|
10
|
-
# def
|
10
|
+
# def revoked
|
11
11
|
# end
|
12
12
|
|
13
13
|
# Define here any teardown things you want when Karafka server stops
|
14
|
-
# def
|
14
|
+
# def shutdown
|
15
15
|
# end
|
16
16
|
end
|
data/lib/karafka/version.rb
CHANGED
data/lib/karafka.rb
CHANGED
data.tar.gz.sig
CHANGED
Binary file
|
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: karafka
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 2.0.0.
|
4
|
+
version: 2.0.0.rc2
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Maciej Mensfeld
|
@@ -34,22 +34,8 @@ cert_chain:
|
|
34
34
|
R2P11bWoCtr70BsccVrN8jEhzwXngMyI2gVt750Y+dbTu1KgRqZKp/ECe7ZzPzXj
|
35
35
|
pIy9vHxTANKYVyI4qj8OrFdEM5BQNu8oQpL0iQ==
|
36
36
|
-----END CERTIFICATE-----
|
37
|
-
date: 2022-
|
37
|
+
date: 2022-07-19 00:00:00.000000000 Z
|
38
38
|
dependencies:
|
39
|
-
- !ruby/object:Gem::Dependency
|
40
|
-
name: dry-configurable
|
41
|
-
requirement: !ruby/object:Gem::Requirement
|
42
|
-
requirements:
|
43
|
-
- - "~>"
|
44
|
-
- !ruby/object:Gem::Version
|
45
|
-
version: '0.13'
|
46
|
-
type: :runtime
|
47
|
-
prerelease: false
|
48
|
-
version_requirements: !ruby/object:Gem::Requirement
|
49
|
-
requirements:
|
50
|
-
- - "~>"
|
51
|
-
- !ruby/object:Gem::Version
|
52
|
-
version: '0.13'
|
53
39
|
- !ruby/object:Gem::Dependency
|
54
40
|
name: dry-monitor
|
55
41
|
requirement: !ruby/object:Gem::Requirement
|
@@ -64,20 +50,6 @@ dependencies:
|
|
64
50
|
- - "~>"
|
65
51
|
- !ruby/object:Gem::Version
|
66
52
|
version: '0.5'
|
67
|
-
- !ruby/object:Gem::Dependency
|
68
|
-
name: dry-validation
|
69
|
-
requirement: !ruby/object:Gem::Requirement
|
70
|
-
requirements:
|
71
|
-
- - "~>"
|
72
|
-
- !ruby/object:Gem::Version
|
73
|
-
version: '1.7'
|
74
|
-
type: :runtime
|
75
|
-
prerelease: false
|
76
|
-
version_requirements: !ruby/object:Gem::Requirement
|
77
|
-
requirements:
|
78
|
-
- - "~>"
|
79
|
-
- !ruby/object:Gem::Version
|
80
|
-
version: '1.7'
|
81
53
|
- !ruby/object:Gem::Dependency
|
82
54
|
name: rdkafka
|
83
55
|
requirement: !ruby/object:Gem::Requirement
|
@@ -112,7 +84,7 @@ dependencies:
|
|
112
84
|
requirements:
|
113
85
|
- - ">="
|
114
86
|
- !ruby/object:Gem::Version
|
115
|
-
version: 2.3.
|
87
|
+
version: 2.3.3
|
116
88
|
- - "<"
|
117
89
|
- !ruby/object:Gem::Version
|
118
90
|
version: 3.0.0
|
@@ -122,7 +94,7 @@ dependencies:
|
|
122
94
|
requirements:
|
123
95
|
- - ">="
|
124
96
|
- !ruby/object:Gem::Version
|
125
|
-
version: 2.3.
|
97
|
+
version: 2.3.3
|
126
98
|
- - "<"
|
127
99
|
- !ruby/object:Gem::Version
|
128
100
|
version: 3.0.0
|
@@ -172,7 +144,8 @@ files:
|
|
172
144
|
- bin/integrations
|
173
145
|
- bin/karafka
|
174
146
|
- bin/scenario
|
175
|
-
- bin/
|
147
|
+
- bin/stress_many
|
148
|
+
- bin/stress_one
|
176
149
|
- bin/wait_for_kafka
|
177
150
|
- certs/karafka-pro.pem
|
178
151
|
- certs/mensfeld.pem
|
@@ -211,6 +184,7 @@ files:
|
|
211
184
|
- lib/karafka/env.rb
|
212
185
|
- lib/karafka/errors.rb
|
213
186
|
- lib/karafka/helpers/async.rb
|
187
|
+
- lib/karafka/helpers/colorize.rb
|
214
188
|
- lib/karafka/helpers/multi_delegator.rb
|
215
189
|
- lib/karafka/instrumentation.rb
|
216
190
|
- lib/karafka/instrumentation/callbacks/error.rb
|
@@ -236,11 +210,15 @@ files:
|
|
236
210
|
- lib/karafka/pro/base_consumer.rb
|
237
211
|
- lib/karafka/pro/loader.rb
|
238
212
|
- lib/karafka/pro/performance_tracker.rb
|
213
|
+
- lib/karafka/pro/processing/coordinator.rb
|
239
214
|
- lib/karafka/pro/processing/jobs/consume_non_blocking.rb
|
240
215
|
- lib/karafka/pro/processing/jobs_builder.rb
|
216
|
+
- lib/karafka/pro/processing/partitioner.rb
|
217
|
+
- lib/karafka/pro/processing/scheduler.rb
|
241
218
|
- lib/karafka/pro/routing/extensions.rb
|
242
|
-
- lib/karafka/pro/scheduler.rb
|
243
219
|
- lib/karafka/process.rb
|
220
|
+
- lib/karafka/processing/coordinator.rb
|
221
|
+
- lib/karafka/processing/coordinators_buffer.rb
|
244
222
|
- lib/karafka/processing/executor.rb
|
245
223
|
- lib/karafka/processing/executors_buffer.rb
|
246
224
|
- lib/karafka/processing/jobs/base.rb
|
@@ -249,7 +227,9 @@ files:
|
|
249
227
|
- lib/karafka/processing/jobs/shutdown.rb
|
250
228
|
- lib/karafka/processing/jobs_builder.rb
|
251
229
|
- lib/karafka/processing/jobs_queue.rb
|
230
|
+
- lib/karafka/processing/partitioner.rb
|
252
231
|
- lib/karafka/processing/result.rb
|
232
|
+
- lib/karafka/processing/scheduler.rb
|
253
233
|
- lib/karafka/processing/worker.rb
|
254
234
|
- lib/karafka/processing/workers_batch.rb
|
255
235
|
- lib/karafka/railtie.rb
|
@@ -263,7 +243,6 @@ files:
|
|
263
243
|
- lib/karafka/routing/topic.rb
|
264
244
|
- lib/karafka/routing/topics.rb
|
265
245
|
- lib/karafka/runner.rb
|
266
|
-
- lib/karafka/scheduler.rb
|
267
246
|
- lib/karafka/serialization/json/deserializer.rb
|
268
247
|
- lib/karafka/server.rb
|
269
248
|
- lib/karafka/setup/config.rb
|
@@ -302,5 +281,5 @@ requirements: []
|
|
302
281
|
rubygems_version: 3.3.7
|
303
282
|
signing_key:
|
304
283
|
specification_version: 4
|
305
|
-
summary: Ruby
|
284
|
+
summary: Ruby framework for working with Apache Kafka
|
306
285
|
test_files: []
|
metadata.gz.sig
CHANGED
Binary file
|
@@ -1,54 +0,0 @@
|
|
1
|
-
# frozen_string_literal: true
|
2
|
-
|
3
|
-
# This Karafka component is a Pro component.
|
4
|
-
# All of the commercial components are present in the lib/karafka/pro directory of this
|
5
|
-
# repository and their usage requires commercial license agreement.
|
6
|
-
#
|
7
|
-
# Karafka has also commercial-friendly license, commercial support and commercial components.
|
8
|
-
#
|
9
|
-
# By sending a pull request to the pro components, you are agreeing to transfer the copyright of
|
10
|
-
# your code to Maciej Mensfeld.
|
11
|
-
|
12
|
-
module Karafka
|
13
|
-
module Pro
|
14
|
-
# Optimizes scheduler that takes into consideration of execution time needed to process
|
15
|
-
# messages from given topics partitions. It uses the non-preemptive LJF algorithm
|
16
|
-
#
|
17
|
-
# This scheduler is designed to optimize execution times on jobs that perform IO operations as
|
18
|
-
# when taking IO into consideration, the can achieve optimized parallel processing.
|
19
|
-
#
|
20
|
-
# This scheduler can also work with virtual partitions.
|
21
|
-
#
|
22
|
-
# Aside from consumption jobs, other jobs do not run often, thus we can leave them with
|
23
|
-
# default FIFO scheduler from the default Karafka scheduler
|
24
|
-
class Scheduler < ::Karafka::Scheduler
|
25
|
-
# Schedules jobs in the LJF order for consumption
|
26
|
-
#
|
27
|
-
# @param queue [Karafka::Processing::JobsQueue] queue where we want to put the jobs
|
28
|
-
# @param jobs_array [Array<Karafka::Processing::Jobs::Base>] jobs we want to schedule
|
29
|
-
#
|
30
|
-
def schedule_consumption(queue, jobs_array)
|
31
|
-
pt = PerformanceTracker.instance
|
32
|
-
|
33
|
-
ordered = []
|
34
|
-
|
35
|
-
jobs_array.each do |job|
|
36
|
-
messages = job.messages
|
37
|
-
message = messages.first
|
38
|
-
|
39
|
-
cost = pt.processing_time_p95(message.topic, message.partition) * messages.size
|
40
|
-
|
41
|
-
ordered << [job, cost]
|
42
|
-
end
|
43
|
-
|
44
|
-
ordered.sort_by!(&:last)
|
45
|
-
ordered.reverse!
|
46
|
-
ordered.map!(&:first)
|
47
|
-
|
48
|
-
ordered.each do |job|
|
49
|
-
queue << job
|
50
|
-
end
|
51
|
-
end
|
52
|
-
end
|
53
|
-
end
|
54
|
-
end
|
data/lib/karafka/scheduler.rb
DELETED
@@ -1,20 +0,0 @@
|
|
1
|
-
# frozen_string_literal: true
|
2
|
-
|
3
|
-
module Karafka
|
4
|
-
# FIFO scheduler for messages coming from various topics and partitions
|
5
|
-
class Scheduler
|
6
|
-
# Schedules jobs in the fifo order
|
7
|
-
#
|
8
|
-
# @param queue [Karafka::Processing::JobsQueue] queue where we want to put the jobs
|
9
|
-
# @param jobs_array [Array<Karafka::Processing::Jobs::Base>] jobs we want to schedule
|
10
|
-
def schedule_consumption(queue, jobs_array)
|
11
|
-
jobs_array.each do |job|
|
12
|
-
queue << job
|
13
|
-
end
|
14
|
-
end
|
15
|
-
|
16
|
-
# Both revocation and shutdown jobs can also run in fifo by default
|
17
|
-
alias schedule_revocation schedule_consumption
|
18
|
-
alias schedule_shutdown schedule_consumption
|
19
|
-
end
|
20
|
-
end
|