skiplock 1.0.9 → 1.0.13
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/README.md +45 -22
- data/bin/skiplock +7 -5
- data/lib/active_job/queue_adapters/skiplock_adapter.rb +1 -1
- data/lib/generators/skiplock/install_generator.rb +1 -1
- data/lib/generators/skiplock/templates/migration.rb.erb +6 -4
- data/lib/skiplock/cron.rb +3 -2
- data/lib/skiplock/extension.rb +25 -0
- data/lib/skiplock/job.rb +49 -30
- data/lib/skiplock/manager.rb +125 -130
- data/lib/skiplock/version.rb +1 -1
- data/lib/skiplock/worker.rb +113 -0
- data/lib/skiplock.rb +17 -16
- metadata +3 -3
- data/lib/skiplock/dispatcher.rb +0 -145
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 89240e4318ca72cf30a3426948d9aa8479f59f85b82e4dca4ef5ee3b449b9894
|
4
|
+
data.tar.gz: b8f6799ccf4ba566379548dc4e9c90b6a1202842139b0ac0a3e0247875be00d9
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 3d8ec5ec2c3b0612fd3077fac5d6ff0c6e1b6826ed74b25b640ccc966874461b470f633c582af4f408f547586e30e1fdc5c22f158ca6fdf5c91000ec33c609a2
|
7
|
+
data.tar.gz: 27b6f4746f255777fdb6b3733d08a0592969a45d6d0f315d76231269255e532b51834fa79e1be89a0c18ac62509fe342266c9a8a5c7c038cc2b9e52b120e2469
|
data/README.md
CHANGED
@@ -48,43 +48,51 @@ The library is quite small compared to other PostgreSQL job queues (eg. *delay_j
|
|
48
48
|
```
|
49
49
|
2. `Skiplock` configuration
|
50
50
|
```yaml
|
51
|
-
# config/skiplock.yml
|
51
|
+
# config/skiplock.yml (default settings)
|
52
52
|
---
|
53
|
-
logging: timestamp
|
54
53
|
min_threads: 1
|
55
54
|
max_threads: 5
|
56
55
|
max_retries: 20
|
57
|
-
|
56
|
+
logfile: log/skiplock.log
|
57
|
+
loglevel: info
|
58
|
+
notification: custom
|
59
|
+
extensions: false
|
58
60
|
purge_completion: true
|
59
61
|
queues:
|
60
62
|
default: 200
|
61
|
-
mailers:
|
63
|
+
mailers: 999
|
62
64
|
workers: 0
|
63
65
|
```
|
64
66
|
Available configuration options are:
|
65
|
-
- **logging** (*enumeration*): sets the logging capability to **true** or **false**; setting to **timestamp** will enable logging with timestamps. The log files are: log/skiplock.log and log/skiplock.error.log
|
66
67
|
- **min_threads** (*integer*): sets minimum number of threads staying idle
|
67
68
|
- **max_threads** (*integer*): sets the maximum number of threads allowed to run jobs
|
68
|
-
- **max_retries** (*integer*): sets the maximum attempt a job will be retrying before it is marked expired
|
69
|
-
- **
|
69
|
+
- **max_retries** (*integer*): sets the maximum attempt a job will be retrying before it is marked expired. See `Retry system` for more details
|
70
|
+
- **logfile** (*string*): path filename for skiplock logs; empty logfile will disable logging
|
71
|
+
- **loglevel** (*string*): sets logging level (`debug, info, warn, error, fatal, unknown`)
|
72
|
+
- **notification** (*string*): sets the library to be used for notifying errors and exceptions (`auto, airbrake, bugsnag, exception_notification, custom`); using `auto` will detect library if available. See `Notification system` for more details
|
73
|
+
- **extensions** (*boolean*): enable or disable the class method extension. See `ClassMethod extension` for more details
|
70
74
|
- **purge_completion** (*boolean*): when set to **true** will delete jobs after they were completed successfully; if set to **false** then the completed jobs should be purged periodically to maximize performance (eg. clean up old jobs after 3 months)
|
71
75
|
- **queues** (*hash*): defines the set of queues with priorities; lower priority takes precedence
|
72
76
|
- **workers** (*integer*) sets the maximum number of processes when running in standalone mode using the `skiplock` executable; setting this to **0** will enable **async mode**
|
73
|
-
|
74
|
-
#### Async mode
|
75
|
-
When **workers** is set to **0** then the jobs will be performed in the web server process using separate threads. If using multi-worker cluster mode web server like Puma, then it should be configured as below:
|
76
|
-
```ruby
|
77
|
-
# config/puma.rb
|
78
|
-
# ...
|
79
|
-
on_worker_fork do |worker_index|
|
80
|
-
Skiplock::Manager.shutdown if worker_index == 1
|
81
|
-
end
|
82
77
|
|
83
|
-
|
84
|
-
|
85
|
-
|
86
|
-
|
78
|
+
#### **Async mode**
|
79
|
+
When **workers** is set to **0** then the jobs will be performed in the web server process using separate threads. If using multi-worker cluster mode web server like Puma, then all the Puma workers will also be able to perform `Skiplock` jobs.
|
80
|
+
|
81
|
+
#### **Standalone mode**
|
82
|
+
`Skiplock` standalone mode can be launched by using the `skiplock` executable; command line options can be provided to override the `Skiplock` configuration file.
|
83
|
+
```
|
84
|
+
$ bundle exec skiplock -h
|
85
|
+
Usage: skiplock [options]
|
86
|
+
-e, --environment STRING Rails environment
|
87
|
+
-l, --logfile STRING Full path to logfile
|
88
|
+
-s, --graceful-shutdown NUM Number of seconds to wait for graceful shutdown
|
89
|
+
-r, --max-retries NUM Number of maxixum retries
|
90
|
+
-t, --max-threads NUM Number of maximum threads
|
91
|
+
-T, --min-threads NUM Number of minimum threads
|
92
|
+
-w, --workers NUM Number of workers
|
93
|
+
-h, --help Show this message
|
87
94
|
```
|
95
|
+
|
88
96
|
## Usage
|
89
97
|
Inside the Rails application:
|
90
98
|
- queue your job
|
@@ -152,17 +160,32 @@ If the retry attempt limit configured in ActiveJob has been reached, then the co
|
|
152
160
|
If the `retry_on` block is not defined, then the built-in retry system of `skiplock` will kick in automatically. The retrying schedule is using an exponential formula (5 + 2**attempt). The `skiplock` configuration `max_retries` determines the the limit of attempts before the failing job is marked as expired. The maximum retry limit can be set as high as 20; this allows up to 12 days of retrying before the job is marked as expired.
|
153
161
|
|
154
162
|
## Notification system
|
155
|
-
`Skiplock` can use existing exception notification library to notify errors and exceptions. It supports `airbrake`, `bugsnag`, and `exception_notification
|
163
|
+
`Skiplock` can use existing exception notification library to notify errors and exceptions. It supports `airbrake`, `bugsnag`, and `exception_notification`. Custom notification can also be called whenever an exception occurs; it can be configured in an initializer like below:
|
156
164
|
```ruby
|
157
165
|
# config/initializers/skiplock.rb
|
158
166
|
Skiplock.on_error do |ex, previous|
|
159
167
|
if ex.backtrace != previous.try(:backtrace)
|
160
168
|
# sends custom email on new exceptions only
|
161
169
|
# the same repeated exceptions will only be sent once to avoid SPAM
|
170
|
+
# NOTE: exceptions generated from Job executions will not provide 'previous' exceptions
|
162
171
|
end
|
163
172
|
end
|
164
|
-
# supports multiple on_error callbacks
|
173
|
+
# supports multiple 'on_error' event callbacks
|
165
174
|
```
|
175
|
+
## ClassMethod extension
|
176
|
+
`Skiplock` can add extension to allow all class methods to be performed as a background job; it is disabled in the default configuration. To enable, edit the `config/skiplock.yml` configuration file and change `extensions` to `true`.
|
177
|
+
- Queue class method `generate_thumbnails` of class `Image` as background job to run as soon as possible
|
178
|
+
```ruby
|
179
|
+
Image.skiplock.generate_thumbnails(height: 100, ratio: true)
|
180
|
+
```
|
181
|
+
- Queue class method `cleanup` of class `Session` as background job on queue `maintenance` to run after 5 minutes
|
182
|
+
```ruby
|
183
|
+
Session.skiplock(wait: 5.minutes, queue: 'maintenance').cleanup
|
184
|
+
```
|
185
|
+
- Queue class method `charge` of class `Subscription` as background job to run tomorrow at noon
|
186
|
+
```ruby
|
187
|
+
Subscription.skiplock(wait_until: Date.tomorrow.noon).charge(amount: 100)
|
188
|
+
```
|
166
189
|
|
167
190
|
## Contributing
|
168
191
|
|
data/bin/skiplock
CHANGED
@@ -5,10 +5,11 @@ begin
|
|
5
5
|
op = OptionParser.new do |opts|
|
6
6
|
opts.banner = "Usage: #{File.basename($0)} [options]"
|
7
7
|
opts.on('-e', '--environment STRING', String, 'Rails environment')
|
8
|
-
opts.on('-l', '--
|
9
|
-
opts.on('-
|
10
|
-
opts.on('-
|
11
|
-
opts.on('-
|
8
|
+
opts.on('-l', '--logfile STRING', String, 'Full path to logfile')
|
9
|
+
opts.on('-s', '--graceful-shutdown NUM', Integer, 'Number of seconds to wait for graceful shutdown')
|
10
|
+
opts.on('-r', '--max-retries NUM', Integer, 'Number of maxixum retries')
|
11
|
+
opts.on('-t', '--max-threads NUM', Integer, 'Number of maximum threads')
|
12
|
+
opts.on('-T', '--min-threads NUM', Integer, 'Number of minimum threads')
|
12
13
|
opts.on('-w', '--workers NUM', Integer, 'Number of workers')
|
13
14
|
opts.on_tail('-h', '--help', 'Show this message') do
|
14
15
|
exit
|
@@ -20,7 +21,8 @@ rescue Exception => e
|
|
20
21
|
puts op
|
21
22
|
exit
|
22
23
|
end
|
24
|
+
options.transform_keys! { |k| k.to_s.gsub('-', '_').to_sym }
|
23
25
|
env = options.delete(:environment)
|
24
26
|
ENV['RAILS_ENV'] = env if env
|
25
27
|
require File.expand_path("config/environment.rb")
|
26
|
-
|
28
|
+
Rails.application.config.skiplock.standalone(**options.merge(standalone: true))
|
@@ -27,10 +27,11 @@ class CreateSkiplockSchema < ActiveRecord::Migration<%= "[#{ActiveRecord::VERSIO
|
|
27
27
|
t.timestamps null: false, default: -> { 'now()' }
|
28
28
|
end
|
29
29
|
create_table 'skiplock.workers', id: :uuid do |t|
|
30
|
-
t.integer :pid, null: false
|
31
|
-
t.integer :
|
30
|
+
t.integer :pid, null: false
|
31
|
+
t.integer :sid, null: false
|
32
32
|
t.integer :capacity, null: false
|
33
33
|
t.string :hostname, null: false, index: true
|
34
|
+
t.boolean :master, null: false, default: false, index: true
|
34
35
|
t.jsonb :data
|
35
36
|
t.timestamps null: false, index: true, default: -> { 'now()' }
|
36
37
|
end
|
@@ -59,7 +60,7 @@ class CreateSkiplockSchema < ActiveRecord::Migration<%= "[#{ActiveRecord::VERSIO
|
|
59
60
|
ELSIF (record.executions IS NOT NULL AND record.scheduled_at IS NOT NULL) THEN
|
60
61
|
INSERT INTO skiplock.counters (day,failures) VALUES (NOW(),1) ON CONFLICT (day) DO UPDATE SET failures = skiplock.counters.failures + 1;
|
61
62
|
END IF;
|
62
|
-
PERFORM pg_notify('skiplock::jobs', CONCAT(TG_OP,',',record.id::TEXT,',',record.worker_id::TEXT,',',record.queue_name,',',record.running::TEXT,',',CAST(EXTRACT(EPOCH FROM record.expired_at) AS FLOAT)::TEXT,',',CAST(EXTRACT(EPOCH FROM record.finished_at) AS FLOAT)::TEXT,',',CAST(EXTRACT(EPOCH FROM CASE WHEN record.scheduled_at IS NULL THEN record.updated_at ELSE record.scheduled_at END) AS FLOAT)::TEXT));
|
63
|
+
PERFORM pg_notify('skiplock::jobs', CONCAT(TG_OP,',',record.id::TEXT,',',record.worker_id::TEXT,',',record.job_class,',',record.queue_name,',',record.running::TEXT,',',CAST(EXTRACT(EPOCH FROM record.expired_at) AS FLOAT)::TEXT,',',CAST(EXTRACT(EPOCH FROM record.finished_at) AS FLOAT)::TEXT,',',CAST(EXTRACT(EPOCH FROM CASE WHEN record.scheduled_at IS NULL THEN record.updated_at ELSE record.scheduled_at END) AS FLOAT)::TEXT));
|
63
64
|
RETURN NULL;
|
64
65
|
END;
|
65
66
|
$$ LANGUAGE plpgsql
|
@@ -74,7 +75,7 @@ class CreateSkiplockSchema < ActiveRecord::Migration<%= "[#{ActiveRecord::VERSIO
|
|
74
75
|
ELSE
|
75
76
|
record = NEW;
|
76
77
|
END IF;
|
77
|
-
PERFORM pg_notify('skiplock::workers', CONCAT(TG_OP,',',record.id::TEXT,',',record.hostname,',',record.capacity,',',record.pid,',',record.
|
78
|
+
PERFORM pg_notify('skiplock::workers', CONCAT(TG_OP,',',record.id::TEXT,',',record.hostname,',',record.master::TEXT,',',record.capacity,',',record.pid,',',record.sid,',',CAST(EXTRACT(EPOCH FROM record.created_at) AS FLOAT)::TEXT,',',CAST(EXTRACT(EPOCH FROM record.updated_at) AS FLOAT)::TEXT));
|
78
79
|
RETURN NULL;
|
79
80
|
END;
|
80
81
|
$$ LANGUAGE plpgsql;
|
@@ -85,6 +86,7 @@ class CreateSkiplockSchema < ActiveRecord::Migration<%= "[#{ActiveRecord::VERSIO
|
|
85
86
|
execute "CREATE INDEX jobs_retry_index ON skiplock.jobs(scheduled_at) WHERE running = FALSE AND executions IS NOT NULL AND expired_at IS NULL AND finished_at IS NULL"
|
86
87
|
execute "CREATE INDEX jobs_cron_index ON skiplock.jobs(scheduled_at ASC NULLS FIRST, priority ASC NULLS LAST, created_at ASC) WHERE cron IS NOT NULL AND finished_at IS NULL"
|
87
88
|
execute "CREATE UNIQUE INDEX jobs_unique_cron_index ON skiplock.jobs (job_class) WHERE cron IS NOT NULL"
|
89
|
+
execute "CREATE UNIQUE INDEX workers_unique_master_index ON skiplock.workers(hostname) WHERE master = 't'"
|
88
90
|
end
|
89
91
|
|
90
92
|
def down
|
data/lib/skiplock/cron.rb
CHANGED
@@ -2,17 +2,18 @@ require 'cron_parser'
|
|
2
2
|
module Skiplock
|
3
3
|
class Cron
|
4
4
|
def self.setup
|
5
|
+
Rails.application.eager_load! if Rails.env.development?
|
5
6
|
cronjobs = []
|
6
7
|
ActiveJob::Base.descendants.each do |j|
|
7
8
|
next unless j.const_defined?('CRON')
|
8
9
|
cron = j.const_get('CRON')
|
9
|
-
job = Job.find_by('job_class = ? AND cron IS NOT NULL', j.name) || Job.new(job_class: j.name, cron: cron)
|
10
|
+
job = Job.find_by('job_class = ? AND cron IS NOT NULL', j.name) || Job.new(job_class: j.name, cron: cron, locale: I18n.locale, timezone: Time.zone.name)
|
10
11
|
time = self.next_schedule_at(cron)
|
11
12
|
if time
|
12
13
|
job.cron = cron
|
13
14
|
job.running = false
|
14
15
|
job.scheduled_at = Time.at(time)
|
15
|
-
job.save
|
16
|
+
job.save
|
16
17
|
cronjobs << j.name
|
17
18
|
end
|
18
19
|
end
|
@@ -0,0 +1,25 @@
|
|
1
|
+
module Skiplock
|
2
|
+
module Extension
|
3
|
+
class Proxy < BasicObject
|
4
|
+
def initialize(target, options = {})
|
5
|
+
@target = target
|
6
|
+
@options = options
|
7
|
+
end
|
8
|
+
|
9
|
+
def method_missing(name, *args)
|
10
|
+
ProxyJob.set(@options).perform_later(::YAML.dump([ @target, name, args ]))
|
11
|
+
end
|
12
|
+
end
|
13
|
+
|
14
|
+
class ProxyJob < ActiveJob::Base
|
15
|
+
def perform(yml)
|
16
|
+
target, method_name, args = ::YAML.load(yml)
|
17
|
+
target.__send__(method_name, *args)
|
18
|
+
end
|
19
|
+
end
|
20
|
+
|
21
|
+
def skiplock(options = {})
|
22
|
+
Proxy.new(self, options)
|
23
|
+
end
|
24
|
+
end
|
25
|
+
end
|
data/lib/skiplock/job.rb
CHANGED
@@ -1,27 +1,6 @@
|
|
1
1
|
module Skiplock
|
2
2
|
class Job < ActiveRecord::Base
|
3
|
-
|
4
|
-
self.connection.exec_query('BEGIN')
|
5
|
-
job = self.find_by_sql("SELECT id, scheduled_at FROM #{self.table_name} WHERE running = FALSE AND expired_at IS NULL AND finished_at IS NULL ORDER BY scheduled_at ASC NULLS FIRST,#{queues_order_query ? ' CASE ' + queues_order_query + ' ELSE NULL END ASC NULLS LAST,' : ''} priority ASC NULLS LAST, created_at ASC FOR UPDATE SKIP LOCKED LIMIT 1").first
|
6
|
-
if job.nil? || job.scheduled_at.to_f > Time.now.to_f
|
7
|
-
self.connection.exec_query('END')
|
8
|
-
return (job ? job.scheduled_at.to_f : Float::INFINITY)
|
9
|
-
end
|
10
|
-
job = Skiplock::Job.find_by_sql("UPDATE #{self.table_name} SET running = TRUE, worker_id = #{self.connection.quote(worker_id)}, updated_at = NOW() WHERE id = '#{job.id}' RETURNING *").first
|
11
|
-
self.connection.exec_query('END')
|
12
|
-
job.data ||= {}
|
13
|
-
job.exception_executions ||= {}
|
14
|
-
job_data = job.attributes.slice('job_class', 'queue_name', 'locale', 'timezone', 'priority', 'executions', 'exception_executions').merge('job_id' => job.id, 'enqueued_at' => job.updated_at, 'arguments' => (job.data['arguments'] || []))
|
15
|
-
job.executions = (job.executions || 0) + 1
|
16
|
-
Thread.current[:skiplock_dispatch_job] = job
|
17
|
-
begin
|
18
|
-
ActiveJob::Base.execute(job_data)
|
19
|
-
rescue Exception => ex
|
20
|
-
end
|
21
|
-
job.dispose(ex)
|
22
|
-
ensure
|
23
|
-
Thread.current[:skiplock_dispatch_job] = nil
|
24
|
-
end
|
3
|
+
self.implicit_order_column = 'created_at'
|
25
4
|
|
26
5
|
def self.enqueue(activejob)
|
27
6
|
self.enqueue_at(activejob, nil)
|
@@ -35,22 +14,28 @@ module Skiplock
|
|
35
14
|
Thread.current[:skiplock_dispatch_job].scheduled_at = timestamp
|
36
15
|
Thread.current[:skiplock_dispatch_job]
|
37
16
|
else
|
38
|
-
|
17
|
+
serialize = activejob.serialize
|
18
|
+
self.create!(serialize.slice(*self.column_names).merge('id' => serialize['job_id'], 'data' => { 'arguments' => serialize['arguments'] }, 'scheduled_at' => timestamp))
|
39
19
|
end
|
40
20
|
end
|
41
21
|
|
42
|
-
def
|
22
|
+
def self.reset_retry_schedules
|
23
|
+
self.where('scheduled_at > NOW() AND executions IS NOT NULL AND expired_at IS NULL AND finished_at IS NULL').update_all(scheduled_at: nil, updated_at: Time.now)
|
24
|
+
end
|
25
|
+
|
26
|
+
def dispose(ex, purge_completion: true, max_retries: 20)
|
27
|
+
yaml = [self, ex].to_yaml
|
43
28
|
self.running = false
|
29
|
+
self.worker_id = nil
|
44
30
|
self.updated_at = (Time.now > self.updated_at ? Time.now : self.updated_at + 1)
|
45
31
|
if ex
|
46
32
|
self.exception_executions["[#{ex.class.name}]"] = (self.exception_executions["[#{ex.class.name}]"] || 0) + 1 unless self.exception_executions.key?('activejob_retry')
|
47
|
-
if self.executions >=
|
33
|
+
if self.executions >= max_retries || self.exception_executions.key?('activejob_retry')
|
48
34
|
self.expired_at = Time.now
|
49
|
-
self.save!
|
50
35
|
else
|
51
36
|
self.scheduled_at = Time.now + (5 * 2**self.executions)
|
52
|
-
self.save!
|
53
37
|
end
|
38
|
+
self.save!
|
54
39
|
Skiplock.on_errors.each { |p| p.call(ex) }
|
55
40
|
elsif self.exception_executions.try(:key?, 'activejob_retry')
|
56
41
|
self.save!
|
@@ -65,9 +50,10 @@ module Skiplock
|
|
65
50
|
self.scheduled_at = Time.at(next_cron_at)
|
66
51
|
self.save!
|
67
52
|
else
|
53
|
+
Skiplock.logger.error "[Skiplock] ERROR: Invalid CRON '#{self.cron}' for Job #{self.job_class}"
|
68
54
|
self.delete
|
69
55
|
end
|
70
|
-
elsif
|
56
|
+
elsif purge_completion
|
71
57
|
self.delete
|
72
58
|
else
|
73
59
|
self.finished_at = Time.now
|
@@ -75,9 +61,42 @@ module Skiplock
|
|
75
61
|
self.save!
|
76
62
|
end
|
77
63
|
self
|
78
|
-
rescue
|
79
|
-
|
64
|
+
rescue Exception => e
|
65
|
+
Skiplock.logger.error(e.name)
|
66
|
+
Skiplock.logger.error(e.backtrace.join("\n"))
|
67
|
+
File.write("tmp/skiplock/#{self.id}", yaml)
|
80
68
|
nil
|
81
69
|
end
|
70
|
+
|
71
|
+
def execute(purge_completion: true, max_retries: 20)
|
72
|
+
Skiplock.logger.info "[Skiplock] Performing #{self.job_class} (#{self.id}) from queue '#{self.queue_name || 'default'}'..."
|
73
|
+
self.data ||= {}
|
74
|
+
self.exception_executions ||= {}
|
75
|
+
job_data = self.attributes.slice('job_class', 'queue_name', 'locale', 'timezone', 'priority', 'executions', 'exception_executions').merge('job_id' => self.id, 'enqueued_at' => self.updated_at, 'arguments' => (self.data['arguments'] || []))
|
76
|
+
self.executions = (self.executions || 0) + 1
|
77
|
+
Thread.current[:skiplock_dispatch_job] = self
|
78
|
+
activejob = ActiveJob::Base.deserialize(job_data)
|
79
|
+
start_time = Process.clock_gettime(Process::CLOCK_MONOTONIC)
|
80
|
+
begin
|
81
|
+
activejob.perform_now
|
82
|
+
rescue Exception => ex
|
83
|
+
end
|
84
|
+
if ex || self.exception_executions.key?('activejob_retry')
|
85
|
+
Skiplock.logger.error("[Skiplock] Job #{self.job_class} (#{self.id}) was interrupted by an exception#{ ' (rescued and retried by ActiveJob)' if self.exception_executions.key?('activejob_retry') }")
|
86
|
+
if ex
|
87
|
+
Skiplock.logger.error(ex)
|
88
|
+
Skiplock.logger.error(ex.backtrace.join("\n"))
|
89
|
+
end
|
90
|
+
else
|
91
|
+
end_time = Process.clock_gettime(Process::CLOCK_MONOTONIC)
|
92
|
+
job_name = self.job_class
|
93
|
+
if self.job_class == 'Skiplock::Extension::ProxyJob'
|
94
|
+
target, method_name = ::YAML.load(self.data['arguments'].first)
|
95
|
+
job_name = "'#{target.name}.#{method_name}'"
|
96
|
+
end
|
97
|
+
Skiplock.logger.info "[Skiplock] Performed #{job_name} (#{self.id}) from queue '#{self.queue_name || 'default'}' in #{end_time - start_time} seconds"
|
98
|
+
end
|
99
|
+
self.dispose(ex, purge_completion: purge_completion, max_retries: max_retries)
|
100
|
+
end
|
82
101
|
end
|
83
102
|
end
|
data/lib/skiplock/manager.rb
CHANGED
@@ -1,62 +1,140 @@
|
|
1
1
|
module Skiplock
|
2
2
|
class Manager
|
3
|
-
def
|
4
|
-
|
5
|
-
|
6
|
-
|
7
|
-
|
8
|
-
|
9
|
-
|
10
|
-
|
11
|
-
|
12
|
-
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
3
|
+
def initialize(**config)
|
4
|
+
@config = Skiplock::DEFAULT_CONFIG.dup
|
5
|
+
@config.merge!(YAML.load_file('config/skiplock.yml')) rescue nil
|
6
|
+
@config.symbolize_keys!
|
7
|
+
@config.transform_values! {|v| v.is_a?(String) ? v.downcase : v}
|
8
|
+
@config.merge!(config)
|
9
|
+
@config[:hostname] = `hostname -f`.strip
|
10
|
+
configure
|
11
|
+
Module.__send__(:include, Skiplock::Extension) if @config[:extensions] == true
|
12
|
+
if (caller.any?{ |l| l =~ %r{/rack/} } && (@config[:workers] == 0 || Rails.env.development?))
|
13
|
+
cleanup_workers
|
14
|
+
@worker = create_worker
|
15
|
+
@thread = @worker.run(**@config)
|
16
|
+
at_exit do
|
17
|
+
@worker.shutdown
|
18
|
+
@thread.join(@config[:graceful_shutdown])
|
19
|
+
@worker.delete
|
20
|
+
end
|
19
21
|
end
|
20
|
-
|
21
|
-
|
22
|
-
|
23
|
-
|
24
|
-
|
25
|
-
|
26
|
-
|
22
|
+
rescue Exception => ex
|
23
|
+
@logger.error(ex.name)
|
24
|
+
@logger.error(ex.backtrace.join("\n"))
|
25
|
+
end
|
26
|
+
|
27
|
+
def standalone(**options)
|
28
|
+
@config.merge!(options)
|
29
|
+
Rails.logger.reopen('/dev/null')
|
30
|
+
Rails.logger.extend(ActiveSupport::Logger.broadcast(@logger))
|
31
|
+
@config[:workers] = 1 if @config[:workers] <= 0
|
32
|
+
banner
|
33
|
+
cleanup_workers
|
34
|
+
@worker = create_worker
|
35
|
+
@parent_id = Process.pid
|
36
|
+
@shutdown = false
|
37
|
+
Signal.trap("INT") { @shutdown = true }
|
38
|
+
Signal.trap("TERM") { @shutdown = true }
|
39
|
+
(@config[:workers] - 1).times do |n|
|
40
|
+
fork do
|
41
|
+
sleep 1
|
42
|
+
worker = create_worker(master: false)
|
43
|
+
thread = worker.run(worker_num: n + 1, **@config)
|
44
|
+
loop do
|
45
|
+
sleep 0.5
|
46
|
+
break if @shutdown || Process.ppid != @parent_id
|
47
|
+
end
|
48
|
+
worker.shutdown
|
49
|
+
thread.join(@config[:graceful_shutdown])
|
50
|
+
worker.delete
|
51
|
+
exit
|
52
|
+
end
|
53
|
+
end
|
54
|
+
@thread = @worker.run(**@config)
|
55
|
+
loop do
|
56
|
+
sleep 0.5
|
57
|
+
break if @shutdown
|
27
58
|
end
|
28
|
-
|
59
|
+
@logger.info "[Skiplock] Terminating signal... Waiting for jobs to finish (up to #{@config[:graceful_shutdown]} seconds)..." if @config[:graceful_shutdown]
|
60
|
+
Process.waitall
|
61
|
+
@worker.shutdown
|
62
|
+
@thread.join(@config[:graceful_shutdown])
|
63
|
+
@worker.delete
|
64
|
+
@logger.info "[Skiplock] Shutdown completed."
|
65
|
+
end
|
66
|
+
|
67
|
+
private
|
68
|
+
|
69
|
+
def banner
|
70
|
+
title = "Skiplock #{Skiplock::VERSION} (Rails #{Rails::VERSION::STRING} | Ruby #{RUBY_VERSION}-p#{RUBY_PATCHLEVEL})"
|
71
|
+
@logger.info "-"*(title.length)
|
72
|
+
@logger.info title
|
73
|
+
@logger.info "-"*(title.length)
|
74
|
+
@logger.info "ClassMethod extensions: #{@config[:extensions]}"
|
75
|
+
@logger.info " Purge completion: #{@config[:purge_completion]}"
|
76
|
+
@logger.info " Notification: #{@config[:notification]}"
|
77
|
+
@logger.info " Max retries: #{@config[:max_retries]}"
|
78
|
+
@logger.info " Min threads: #{@config[:min_threads]}"
|
79
|
+
@logger.info " Max threads: #{@config[:max_threads]}"
|
80
|
+
@logger.info " Environment: #{Rails.env}"
|
81
|
+
@logger.info " Loglevel: #{@config[:loglevel]}"
|
82
|
+
@logger.info " Logfile: #{@config[:logfile] || '(disabled)'}"
|
83
|
+
@logger.info " Workers: #{@config[:workers]}"
|
84
|
+
@logger.info " Queues: #{@config[:queues].map {|k,v| k + '(' + v.to_s + ')'}.join(', ')}" if @config[:queues].is_a?(Hash)
|
85
|
+
@logger.info " PID: #{Process.pid}"
|
86
|
+
@logger.info "-"*(title.length)
|
87
|
+
@logger.warn "[Skiplock] Custom notification has no registered 'on_error' callback" if Skiplock.on_errors.count == 0
|
29
88
|
end
|
30
|
-
|
31
|
-
def
|
32
|
-
|
33
|
-
|
34
|
-
|
35
|
-
|
36
|
-
@thread = nil
|
89
|
+
|
90
|
+
def cleanup_workers
|
91
|
+
delete_ids = []
|
92
|
+
Worker.where(hostname: @config[:hostname]).each do |worker|
|
93
|
+
sid = Process.getsid(worker.pid) rescue nil
|
94
|
+
delete_ids << worker.id if worker.sid != sid || worker.updated_at < 30.minutes.ago
|
37
95
|
end
|
96
|
+
Worker.where(id: delete_ids).delete_all if delete_ids.count > 0
|
97
|
+
Job.where(running: true).where.not(worker_id: Worker.ids).update_all(running: false, worker_id: nil)
|
38
98
|
end
|
39
99
|
|
40
|
-
|
100
|
+
def create_worker(master: true)
|
101
|
+
Worker.create!(pid: Process.pid, sid: Process.getsid(), master: master, hostname: @config[:hostname], capacity: @config[:max_threads])
|
102
|
+
rescue
|
103
|
+
Worker.create!(pid: Process.pid, sid: Process.getsid(), master: false, hostname: @config[:hostname], capacity: @config[:max_threads])
|
104
|
+
end
|
41
105
|
|
42
|
-
def
|
43
|
-
config =
|
44
|
-
|
45
|
-
|
46
|
-
@
|
47
|
-
if @
|
106
|
+
def configure
|
107
|
+
@config[:loglevel] = 'info' unless ['debug','info','warn','error','fatal','unknown'].include?(@config[:loglevel].to_s)
|
108
|
+
@config[:graceful_shutdown] = 300 if @config[:graceful_shutdown] > 300
|
109
|
+
@config[:graceful_shutdown] = nil if @config[:graceful_shutdown] < 0
|
110
|
+
@config[:max_retries] = 20 if @config[:max_retries] > 20
|
111
|
+
@config[:max_retries] = 0 if @config[:max_retries] < 0
|
112
|
+
@config[:max_threads] = 1 if @config[:max_threads] < 1
|
113
|
+
@config[:max_threads] = 20 if @config[:max_threads] > 20
|
114
|
+
@config[:min_threads] = 0 if @config[:min_threads] < 0
|
115
|
+
@config[:workers] = 0 if @config[:workers] < 0
|
116
|
+
@logger = ActiveSupport::Logger.new(STDOUT)
|
117
|
+
@logger.level = @config[:loglevel].to_sym
|
118
|
+
Skiplock.logger = @logger
|
119
|
+
raise "Cannot create logfile '#{@config[:logfile]}'" if @config[:logfile] && !File.writable?(File.dirname(@config[:logfile]))
|
120
|
+
@config[:logfile] = nil if @config[:logfile].to_s.length == 0
|
121
|
+
if @config[:logfile]
|
122
|
+
@logger.extend(ActiveSupport::Logger.broadcast(::Logger.new(@config[:logfile])))
|
123
|
+
ActiveJob::Base.logger = nil
|
124
|
+
end
|
125
|
+
@config[:queues].values.each { |v| raise 'Queue value must be an integer' unless v.is_a?(Integer) } if @config[:queues].is_a?(Hash)
|
126
|
+
if @config[:notification] == 'auto'
|
48
127
|
if defined?(Airbrake)
|
49
|
-
@notification = 'airbrake'
|
128
|
+
@config[:notification] = 'airbrake'
|
50
129
|
elsif defined?(Bugsnag)
|
51
|
-
@notification = 'bugsnag'
|
130
|
+
@config[:notification] = 'bugsnag'
|
52
131
|
elsif defined?(ExceptionNotifier)
|
53
|
-
@notification = 'exception_notification'
|
132
|
+
@config[:notification] = 'exception_notification'
|
54
133
|
else
|
55
|
-
|
56
|
-
exit
|
134
|
+
raise "Unable to detect any known exception notification library. Please define custom 'on_error' event callbacks and change to 'custom' notification in 'config/skiplock.yml'"
|
57
135
|
end
|
58
136
|
end
|
59
|
-
case @notification
|
137
|
+
case @config[:notification]
|
60
138
|
when 'airbrake'
|
61
139
|
raise 'airbrake gem not found' unless defined?(Airbrake)
|
62
140
|
Skiplock.on_error do |ex, previous|
|
@@ -72,93 +150,10 @@ module Skiplock
|
|
72
150
|
Skiplock.on_error do |ex, previous|
|
73
151
|
ExceptionNotifier.notify_exception(ex) unless ex.backtrace == previous.try(:backtrace)
|
74
152
|
end
|
153
|
+
else
|
154
|
+
@config[:notification] = 'custom'
|
75
155
|
end
|
76
|
-
|
77
|
-
STDERR.puts "Invalid configuration 'config/skiplock.yml': #{e.message}"
|
78
|
-
exit
|
79
|
-
end
|
80
|
-
|
81
|
-
def self.standalone
|
82
|
-
if Settings['logging']
|
83
|
-
log_timestamp = (Settings['logging'].to_s == 'timestamp')
|
84
|
-
logfile = File.open('log/skiplock.log', 'a')
|
85
|
-
logfile.sync = true
|
86
|
-
$stdout = Demux.new(logfile, STDOUT, timestamp: log_timestamp)
|
87
|
-
errfile = File.open('log/skiplock.error.log', 'a')
|
88
|
-
errfile.sync = true
|
89
|
-
$stderr = Demux.new(errfile, STDERR, timestamp: log_timestamp)
|
90
|
-
logger = ActiveSupport::Logger.new($stdout)
|
91
|
-
logger.level = Rails.logger.level
|
92
|
-
Rails.logger.reopen('/dev/null')
|
93
|
-
Rails.logger.extend(ActiveSupport::Logger.broadcast(logger))
|
94
|
-
end
|
95
|
-
title = "Skiplock version: #{Skiplock::VERSION} (Ruby #{RUBY_VERSION}-p#{RUBY_PATCHLEVEL})"
|
96
|
-
puts "-"*(title.length)
|
97
|
-
puts title
|
98
|
-
puts "-"*(title.length)
|
99
|
-
puts "Purge completion: #{Settings['purge_completion']}"
|
100
|
-
puts " Notification: #{Settings['notification']}#{(' (' + @notification + ')') if Settings['notification'] == 'auto'}"
|
101
|
-
puts " Max retries: #{Settings['max_retries']}"
|
102
|
-
puts " Min threads: #{Settings['min_threads']}"
|
103
|
-
puts " Max threads: #{Settings['max_threads']}"
|
104
|
-
puts " Environment: #{Rails.env}"
|
105
|
-
puts " Logging: #{Settings['logging']}"
|
106
|
-
puts " Workers: #{Settings['workers']}"
|
107
|
-
puts " Queues: #{Settings['queues'].map {|k,v| k + '(' + v.to_s + ')'}.join(', ')}" if Settings['queues'].is_a?(Hash)
|
108
|
-
puts " PID: #{Process.pid}"
|
109
|
-
puts "-"*(title.length)
|
110
|
-
parent_id = Process.pid
|
111
|
-
shutdown = false
|
112
|
-
Signal.trap("INT") { shutdown = true }
|
113
|
-
Signal.trap("TERM") { shutdown = true }
|
114
|
-
worker_pids = []
|
115
|
-
(Settings['workers']-1).times do |n|
|
116
|
-
worker_pids << fork do
|
117
|
-
dispatcher = Dispatcher.new(master: false, worker_num: n+1)
|
118
|
-
thread = dispatcher.run
|
119
|
-
loop do
|
120
|
-
sleep 0.5
|
121
|
-
break if shutdown || Process.ppid != parent_id
|
122
|
-
end
|
123
|
-
dispatcher.shutdown(wait: true)
|
124
|
-
thread.join
|
125
|
-
exit
|
126
|
-
end
|
127
|
-
end
|
128
|
-
sleep 0.1
|
129
|
-
dispatcher = Dispatcher.new(worker_pids: worker_pids)
|
130
|
-
thread = dispatcher.run
|
131
|
-
loop do
|
132
|
-
sleep 0.5
|
133
|
-
break if shutdown
|
134
|
-
end
|
135
|
-
Process.waitall
|
136
|
-
dispatcher.shutdown(wait: true)
|
137
|
-
thread.join
|
138
|
-
end
|
139
|
-
|
140
|
-
class Demux
|
141
|
-
def initialize(*targets, timestamp: true)
|
142
|
-
@targets = targets
|
143
|
-
@timestamp = timestamp
|
144
|
-
end
|
145
|
-
|
146
|
-
def close
|
147
|
-
@targets.each(&:close)
|
148
|
-
end
|
149
|
-
|
150
|
-
def flush
|
151
|
-
@targets.each(&:flush)
|
152
|
-
end
|
153
|
-
|
154
|
-
def tty?
|
155
|
-
true
|
156
|
-
end
|
157
|
-
|
158
|
-
def write(*args)
|
159
|
-
args.prepend("[#{Time.now.utc}]: ") if @timestamp
|
160
|
-
@targets.each {|t| t.write(*args)}
|
161
|
-
end
|
156
|
+
Skiplock.on_errors.freeze unless Skiplock.on_errors.frozen?
|
162
157
|
end
|
163
158
|
end
|
164
159
|
end
|
data/lib/skiplock/version.rb
CHANGED
data/lib/skiplock/worker.rb
CHANGED
@@ -1,4 +1,117 @@
|
|
1
1
|
module Skiplock
|
2
2
|
class Worker < ActiveRecord::Base
|
3
|
+
self.implicit_order_column = 'created_at'
|
4
|
+
|
5
|
+
def run(worker_num: 0, **config)
|
6
|
+
@config = config
|
7
|
+
@worker_num = worker_num
|
8
|
+
@queues_order_query = @config[:queues].map { |q,v| "WHEN queue_name = '#{q}' THEN #{v}" }.join(' ') if @config[:queues].is_a?(Hash) && @config[:queues].count > 0
|
9
|
+
@next_schedule_at = Time.now.to_f
|
10
|
+
@executor = Concurrent::ThreadPoolExecutor.new(min_threads: @config[:min_threads], max_threads: @config[:max_threads], max_queue: @config[:max_threads], idletime: 60, auto_terminate: true, fallback_policy: :discard)
|
11
|
+
@running = true
|
12
|
+
Process.setproctitle("skiplock-#{self.master ? 'master[0]' : 'worker[' + @worker_num.to_s + ']'}") if @config[:standalone]
|
13
|
+
Thread.new do
|
14
|
+
@connection = self.class.connection
|
15
|
+
@connection.exec_query('LISTEN "skiplock::jobs"')
|
16
|
+
if self.master
|
17
|
+
Dir.mkdir('tmp/skiplock') unless Dir.exist?('tmp/skiplock')
|
18
|
+
check_sync_errors
|
19
|
+
Cron.setup
|
20
|
+
end
|
21
|
+
error = false
|
22
|
+
timestamp = Process.clock_gettime(Process::CLOCK_MONOTONIC)
|
23
|
+
while @running
|
24
|
+
Rails.application.reloader.wrap do
|
25
|
+
begin
|
26
|
+
if error
|
27
|
+
unless @connection.active?
|
28
|
+
@connection.reconnect!
|
29
|
+
sleep(0.5)
|
30
|
+
@connection.exec_query('LISTEN "skiplock::jobs"')
|
31
|
+
@next_schedule_at = Time.now.to_f
|
32
|
+
end
|
33
|
+
check_sync_errors if self.master
|
34
|
+
error = false
|
35
|
+
end
|
36
|
+
job_notifications = []
|
37
|
+
@connection.raw_connection.wait_for_notify(0.1) do |channel, pid, payload|
|
38
|
+
job_notifications << payload if payload
|
39
|
+
loop do
|
40
|
+
payload = @connection.raw_connection.notifies
|
41
|
+
break unless @running && payload
|
42
|
+
job_notifications << payload[:extra]
|
43
|
+
end
|
44
|
+
job_notifications.each do |n|
|
45
|
+
op, id, worker_id, job_class, queue_name, running, expired_at, finished_at, scheduled_at = n.split(',')
|
46
|
+
next if op == 'DELETE' || running == 'true' || expired_at.to_f > 0 || finished_at.to_f > 0
|
47
|
+
if scheduled_at.to_f <= Time.now.to_f
|
48
|
+
@next_schedule_at = Time.now.to_f
|
49
|
+
elsif scheduled_at.to_f < @next_schedule_at
|
50
|
+
@next_schedule_at = scheduled_at.to_f
|
51
|
+
end
|
52
|
+
end
|
53
|
+
end
|
54
|
+
if Time.now.to_f >= @next_schedule_at && @executor.remaining_capacity > 0
|
55
|
+
job = dispatch_job
|
56
|
+
if job.is_a?(Job)
|
57
|
+
@executor.post(job, @config[:purge_completion], @config[:max_retries]) do |job, purge_completion, max_retries|
|
58
|
+
job.execute(purge_completion: purge_completion, max_retries: max_retries)
|
59
|
+
end
|
60
|
+
else
|
61
|
+
@next_schedule_at = job
|
62
|
+
end
|
63
|
+
end
|
64
|
+
if Process.clock_gettime(Process::CLOCK_MONOTONIC) - timestamp > 60
|
65
|
+
self.touch
|
66
|
+
timestamp = Process.clock_gettime(Process::CLOCK_MONOTONIC)
|
67
|
+
end
|
68
|
+
rescue Exception => ex
|
69
|
+
# most likely error with database connection
|
70
|
+
Skiplock.logger.error(ex.name)
|
71
|
+
Skiplock.logger.error(ex.backtrace.join("\n"))
|
72
|
+
Skiplock.on_errors.each { |p| p.call(ex, @last_exception) }
|
73
|
+
error = true
|
74
|
+
t = Process.clock_gettime(Process::CLOCK_MONOTONIC)
|
75
|
+
while @running
|
76
|
+
sleep(0.5)
|
77
|
+
break if Process.clock_gettime(Process::CLOCK_MONOTONIC) - t > 5
|
78
|
+
end
|
79
|
+
@last_exception = ex
|
80
|
+
end
|
81
|
+
sleep(0.2)
|
82
|
+
end
|
83
|
+
end
|
84
|
+
@connection.exec_query('UNLISTEN *')
|
85
|
+
@executor.shutdown
|
86
|
+
@executor.kill unless @executor.wait_for_termination(@config[:graceful_shutdown])
|
87
|
+
end
|
88
|
+
end
|
89
|
+
|
90
|
+
def shutdown
|
91
|
+
@running = false
|
92
|
+
end
|
93
|
+
|
94
|
+
private
|
95
|
+
|
96
|
+
def check_sync_errors
|
97
|
+
# get performed jobs that could not sync with database
|
98
|
+
Dir.glob('tmp/skiplock/*').each do |f|
|
99
|
+
job_from_db = Job.find_by(id: File.basename(f), running: true)
|
100
|
+
disposed = true
|
101
|
+
if job_from_db
|
102
|
+
job, ex = YAML.load_file(f) rescue nil
|
103
|
+
disposed = job.dispose(ex, purge_completion: @config[:purge_completion], max_retries: @config[:max_retries])
|
104
|
+
end
|
105
|
+
File.delete(f) if disposed
|
106
|
+
end
|
107
|
+
end
|
108
|
+
|
109
|
+
def dispatch_job
|
110
|
+
@connection.transaction do
|
111
|
+
job = Job.find_by_sql("SELECT id, scheduled_at FROM skiplock.jobs WHERE running = FALSE AND expired_at IS NULL AND finished_at IS NULL ORDER BY scheduled_at ASC NULLS FIRST,#{@queues_order_query ? ' CASE ' + @queues_order_query + ' ELSE NULL END ASC NULLS LAST,' : ''} priority ASC NULLS LAST, created_at ASC FOR UPDATE SKIP LOCKED LIMIT 1").first
|
112
|
+
return (job ? job.scheduled_at.to_f : Float::INFINITY) if job.nil? || job.scheduled_at.to_f > Time.now.to_f
|
113
|
+
Job.find_by_sql("UPDATE skiplock.jobs SET running = TRUE, worker_id = '#{self.id}', updated_at = NOW() WHERE id = '#{job.id}' RETURNING *").first
|
114
|
+
end
|
115
|
+
end
|
3
116
|
end
|
4
117
|
end
|
data/lib/skiplock.rb
CHANGED
@@ -3,30 +3,31 @@ require 'active_job/queue_adapters/skiplock_adapter'
|
|
3
3
|
require 'active_record'
|
4
4
|
require 'skiplock/counter'
|
5
5
|
require 'skiplock/cron'
|
6
|
-
require 'skiplock/
|
6
|
+
require 'skiplock/extension'
|
7
7
|
require 'skiplock/job'
|
8
8
|
require 'skiplock/manager'
|
9
9
|
require 'skiplock/worker'
|
10
10
|
require 'skiplock/version'
|
11
11
|
|
12
12
|
module Skiplock
|
13
|
-
|
14
|
-
|
15
|
-
|
16
|
-
|
17
|
-
|
18
|
-
|
19
|
-
|
20
|
-
|
21
|
-
|
22
|
-
'mailers' => 100
|
23
|
-
},
|
24
|
-
'workers' => 0
|
25
|
-
}
|
26
|
-
mattr_reader :on_errors, default: []
|
13
|
+
DEFAULT_CONFIG = { 'extensions' => false, 'logfile' => 'log/skiplock.log', 'loglevel' => 'info', 'graceful_shutdown' => 15, 'min_threads' => 1, 'max_threads' => 5, 'max_retries' => 20, 'notification' => 'custom', 'purge_completion' => true, 'queues' => { 'default' => 100, 'mailers' => 999 }, 'workers' => 0 }.freeze
|
14
|
+
|
15
|
+
def self.logger=(l)
|
16
|
+
@logger = l
|
17
|
+
end
|
18
|
+
|
19
|
+
def self.logger
|
20
|
+
@logger
|
21
|
+
end
|
27
22
|
|
28
23
|
def self.on_error(&block)
|
29
|
-
|
24
|
+
@on_errors ||= []
|
25
|
+
@on_errors << block
|
26
|
+
block
|
27
|
+
end
|
28
|
+
|
29
|
+
def self.on_errors
|
30
|
+
@on_errors || []
|
30
31
|
end
|
31
32
|
|
32
33
|
def self.table_name_prefix
|
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: skiplock
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.0.
|
4
|
+
version: 1.0.13
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Tin Vo
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2021-
|
11
|
+
date: 2021-09-04 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: activejob
|
@@ -84,7 +84,7 @@ files:
|
|
84
84
|
- lib/skiplock.rb
|
85
85
|
- lib/skiplock/counter.rb
|
86
86
|
- lib/skiplock/cron.rb
|
87
|
-
- lib/skiplock/
|
87
|
+
- lib/skiplock/extension.rb
|
88
88
|
- lib/skiplock/job.rb
|
89
89
|
- lib/skiplock/manager.rb
|
90
90
|
- lib/skiplock/version.rb
|
data/lib/skiplock/dispatcher.rb
DELETED
@@ -1,145 +0,0 @@
|
|
1
|
-
module Skiplock
|
2
|
-
class Dispatcher
|
3
|
-
def initialize(master: true, worker_num: nil, worker_pids: [])
|
4
|
-
@queues_order_query = Settings['queues'].map { |q,v| "WHEN queue_name = '#{q}' THEN #{v}" }.join(' ') if Settings['queues'].is_a?(Hash) && Settings['queues'].count > 0
|
5
|
-
@executor = Concurrent::ThreadPoolExecutor.new(min_threads: Settings['min_threads'], max_threads: Settings['max_threads'], max_queue: Settings['max_threads'], idletime: 60, auto_terminate: true, fallback_policy: :discard)
|
6
|
-
@master = master
|
7
|
-
if @master
|
8
|
-
@worker_pids = worker_pids + [ Process.pid ]
|
9
|
-
else
|
10
|
-
@worker_num = worker_num
|
11
|
-
end
|
12
|
-
@last_dispatch_at = 0
|
13
|
-
@next_schedule_at = Time.now.to_f
|
14
|
-
@running = true
|
15
|
-
end
|
16
|
-
|
17
|
-
def run
|
18
|
-
Thread.new do
|
19
|
-
Rails.application.reloader.wrap do
|
20
|
-
sleep(1) while @running && !Rails.application.initialized?
|
21
|
-
Rails.application.eager_load!
|
22
|
-
Process.setproctitle("skiplock-#{@master ? 'master[0]' : 'worker[' + @worker_num.to_s + ']'}") if Settings['workers'] > 0 && !Rails.env.development?
|
23
|
-
ActiveRecord::Base.connection_pool.with_connection do |connection|
|
24
|
-
connection.exec_query('LISTEN "skiplock::jobs"')
|
25
|
-
hostname = `hostname -f`.strip
|
26
|
-
@worker = Worker.create!(pid: Process.pid, ppid: (@master ? nil : Process.ppid), capacity: Settings['max_threads'], hostname: hostname)
|
27
|
-
if @master
|
28
|
-
connection.exec_query('LISTEN "skiplock::workers"')
|
29
|
-
Dir.mkdir('tmp/skiplock') unless Dir.exist?('tmp/skiplock')
|
30
|
-
check_sync_errors
|
31
|
-
# get dead worker ids
|
32
|
-
dead_worker_ids = Worker.where(hostname: hostname).where.not(pid: @worker_pids).ids
|
33
|
-
if dead_worker_ids.count > 0
|
34
|
-
# reset orphaned jobs of the dead worker ids for retry
|
35
|
-
Job.where(running: true).where(worker_id: dead_worker_ids).update_all(running: false, worker_id: nil)
|
36
|
-
# remove dead workers
|
37
|
-
Worker.where(id: dead_worker_ids).delete_all
|
38
|
-
end
|
39
|
-
# reset retries schedules on startup
|
40
|
-
Job.where('scheduled_at > NOW() AND executions IS NOT NULL AND expired_at IS NULL AND finished_at IS NULL').update_all(scheduled_at: nil, updated_at: Time.now)
|
41
|
-
Cron.setup
|
42
|
-
end
|
43
|
-
error = false
|
44
|
-
while @running
|
45
|
-
begin
|
46
|
-
if error
|
47
|
-
unless connection.active?
|
48
|
-
connection.reconnect!
|
49
|
-
sleep(0.5)
|
50
|
-
connection.exec_query('LISTEN "skiplock::jobs"')
|
51
|
-
connection.exec_query('LISTEN "skiplock::workers"') if @master
|
52
|
-
@next_schedule_at = Time.now.to_f
|
53
|
-
end
|
54
|
-
check_sync_errors
|
55
|
-
error = false
|
56
|
-
end
|
57
|
-
notifications = { 'skiplock::jobs' => [], 'skiplock::workers' => [] }
|
58
|
-
connection.raw_connection.wait_for_notify(0.1) do |channel, pid, payload|
|
59
|
-
notifications[channel] << payload if payload
|
60
|
-
loop do
|
61
|
-
payload = connection.raw_connection.notifies
|
62
|
-
break unless @running && payload
|
63
|
-
notifications[payload[:relname]] << payload[:extra]
|
64
|
-
end
|
65
|
-
notifications['skiplock::jobs'].each do |n|
|
66
|
-
op, id, worker_id, queue_name, running, expired_at, finished_at, scheduled_at = n.split(',')
|
67
|
-
if @master
|
68
|
-
# TODO: report job status to action cable
|
69
|
-
end
|
70
|
-
next if op == 'DELETE' || running == 'true' || expired_at.to_f > 0 || finished_at.to_f > 0 || scheduled_at.to_f < @last_dispatch_at
|
71
|
-
if scheduled_at.to_f < Time.now.to_f
|
72
|
-
@next_schedule_at = Time.now.to_f
|
73
|
-
elsif scheduled_at.to_f < @next_schedule_at
|
74
|
-
@next_schedule_at = scheduled_at.to_f
|
75
|
-
end
|
76
|
-
end
|
77
|
-
if @master
|
78
|
-
# TODO: report worker status to action cable
|
79
|
-
notifications['skiplock::workers'].each do |n|
|
80
|
-
end
|
81
|
-
end
|
82
|
-
end
|
83
|
-
if Time.now.to_f >= @next_schedule_at && @executor.remaining_capacity > 0
|
84
|
-
@executor.post { do_work }
|
85
|
-
end
|
86
|
-
rescue Exception => ex
|
87
|
-
# most likely error with database connection
|
88
|
-
STDERR.puts ex.message
|
89
|
-
STDERR.puts ex.backtrace
|
90
|
-
Skiplock.on_errors.each { |p| p.call(ex, @last_exception) }
|
91
|
-
error = true
|
92
|
-
t = Time.now
|
93
|
-
while @running
|
94
|
-
sleep(0.5)
|
95
|
-
break if Time.now - t > 5
|
96
|
-
end
|
97
|
-
@last_exception = ex
|
98
|
-
end
|
99
|
-
sleep(0.2)
|
100
|
-
end
|
101
|
-
connection.exec_query('UNLISTEN *')
|
102
|
-
@executor.shutdown
|
103
|
-
@executor.wait_for_termination if @wait
|
104
|
-
@worker.delete if @worker
|
105
|
-
end
|
106
|
-
end
|
107
|
-
end
|
108
|
-
end
|
109
|
-
|
110
|
-
def shutdown(wait: true)
|
111
|
-
@running = false
|
112
|
-
@wait = wait
|
113
|
-
end
|
114
|
-
|
115
|
-
private
|
116
|
-
|
117
|
-
def check_sync_errors
|
118
|
-
# get performed jobs that could not sync with database
|
119
|
-
Dir.glob('tmp/skiplock/*').each do |f|
|
120
|
-
job_from_db = Job.find_by(id: File.basename(f), running: true)
|
121
|
-
disposed = true
|
122
|
-
if job_from_db
|
123
|
-
job, ex = YAML.load_file(f) rescue nil
|
124
|
-
disposed = job.dispose(ex)
|
125
|
-
end
|
126
|
-
File.delete(f) if disposed
|
127
|
-
end
|
128
|
-
end
|
129
|
-
|
130
|
-
def do_work
|
131
|
-
while @running
|
132
|
-
@last_dispatch_at = Time.now.to_f - 1 # 1 second allowance for timedrift
|
133
|
-
result = Job.dispatch(queues_order_query: @queues_order_query, worker_id: @worker.id)
|
134
|
-
next if result.is_a?(Job)
|
135
|
-
@next_schedule_at = result if result.is_a?(Float)
|
136
|
-
break
|
137
|
-
end
|
138
|
-
rescue Exception => ex
|
139
|
-
STDERR.puts ex.message
|
140
|
-
STDERR.puts ex.backtrace
|
141
|
-
Skiplock.on_errors.each { |p| p.call(ex, @last_exception) }
|
142
|
-
@last_exception = ex
|
143
|
-
end
|
144
|
-
end
|
145
|
-
end
|