-
Notifications
You must be signed in to change notification settings - Fork 4
/
README
308 lines (222 loc) · 10.8 KB
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
NAME
bj
SYNOPSIS
bj (migration_code|generate_migration|migrate|setup|plugin|run|submit|list|set|config|pid) [options]+
DESCRIPTION
________________________________
Overview
--------------------------------
Backgroundjob (Bj) is a brain dead simple zero admin background priority queue
for Rails. Bj is robust, platform independent (including windows), and
supports internal or external manangement of the background runner process.
Jobs can be submitted to the queue directly using the api or from the command
line using the ./script/bj:
api:
Bj.submit 'cat /etc/password'
command line:
bj submit cat /etc/password
Bj's priority queue lives in the database and is therefore durable - your jobs
will live across an app crash or machine reboot. The job management is
comprehensive capturing stdout, stderr, exit_status, and temporal statistics
about each job:
jobs = Bj.submit array_of_commands, :priority => 42
...
jobs.each do |job|
if job.finished?
p job.stdout
p job.stderr
p job.exit_status
p job.started_at
p job.finished_at
end
end
In addition the background runner process logs all commands run and their
exit_status to a log named using the following convention:
rails_root/log/bj.#{ HOSTNAME }.#{ RAILS_ENV }.log
Bj allows you to submit jobs to multiple databases; for instance, if your
application is running in development mode you may do:
Bj.in :production do
Bj.submit 'my_job.exe'
end
Bj manages the ever growing list of jobs ran by automatically archiving them
into another table (by default jobs > 24 hrs old are archived) to prevent the
jobs table from becoming bloated and huge.
All Bj's tables are namespaced and accessible via the Bj module:
Bj.table.job.find(:all) # jobs table
Bj.table.job_archive.find(:all) # archived jobs
Bj.table.config.find(:all) # configuration and runner state
Bj always arranges for submitted jobs to run with a current working directory
of RAILS_ROOT and with the correct RAILS_ENV setting. For example, if you
submit a job in production it will have ENV['RAILS_ENV'] == 'production'.
When Bj manages the background runner it will never outlive the rails
application - it is started and stopped on demand as the rails app is started
and stopped. This is also true for ./script/console - Bj will automatically
fire off the background runner to process jobs submitted using the console.
Bj ensures that only one background process is running for your application -
firing up three mongrels or fcgi processes will result in only one background
runner being started. Note that the number of background runners does not
determine throughput - that is determined primarily by the nature of the jobs
themselves and how much work they perform per process.
________________________________
Architecture
--------------------------------
If one ignores platform specific details the design of Bj is quite simple: the
main Rails application submits jobs to table, stored in the database. The act
of submitting triggers exactly one of two things to occur:
1) a new long running background runner to be started
2) an existing background runner to be signaled
The background runner refuses to run two copies of itself for a given
hostname/rails_env combination. For example you may only have one background
runner processing jobs on localhost in development mode.
The background runner, under normal circumstances, is managed by Bj itself -
you need do nothing to start, monitor, or stop it - it just works. However,
some people will prefer manage their own background process, see 'External
Runner' section below for more on this.
The runner simply processes each job in a highest priority oldest-in fashion,
capturing stdout, stderr, exit_status, etc. and storing the information back
into the database while logging it's actions. When there are no jobs to run
the runner goes to sleep for 42 seconds; however this sleep is interuptable,
such as when the runner is signaled that a new job has been submitted so,
under normal circumstances there will be zero lag between job submission and
job running for an empty queue.
________________________________
External Runner / Clustering
--------------------------------
For the paranoid control freaks out there (myself included) it is quite
possible to manage and monitor the runner process manually. This can be
desirable in production setups where monitoring software may kill leaking
rails apps periodically.
Recalling that Bj will only allow one copy of itself to process jobs per
hostname/rails_env pair we can simply do something like this in cron
cmd = bj run --forever \
--rails_env=development \
--rails_root=/Users/ahoward/rails_root
*/15 * * * * $cmd
this will simply attempt the start the background runner every 15 minutes if,
and only if, it's not *already* running.
In addtion to this you'll want to tell Bj not to manage the runner itself
using
Bj.config["production.no_tickle"] = true
Note that, for clusting setups, it's as simple as adding a crontab and config
entry like this for each host. Because Bj throttles background runners per
hostname this will allow one runner per hostname - making it quite simple to
cluster three nodes behind a besieged rails application.
________________________________
Designing Jobs
--------------------------------
Bj runs it's jobs as command line applications. It ensures that all jobs run
in RAILS_ROOT so it's quite natural to apply a pattern such as
mkdir ./jobs
edit ./jobs/background_job_to_run
...
Bj.submit "./jobs/background_job_to_run"
If you need to run you jobs under an entire rails environment you'll need to
do this:
Bj.submit "./script/runner ./jobs/background_job_to_run"
Obviously "./script/runner" loads the rails environment for you. It's worth
noting that this happens for each job and that this is by design: the reason
is that most rails applications leak memory like a sieve so, if one were to
spawn a long running process that used the application code base you'd have a
lovely doubling of memory usage on you app servers. Although loading the
rails environment for each background job requires a little time, a little
cpu, and a lot less memory. A future version of Bj will provide a way to load
the rails environment once and to process background jobs in this environment,
but anyone wanting to use this in production will be required to duct tape
their entire chest and have a team of oxen rip off the tape without screaming
to prove steelyness of spirit and profound understanding of the other side.
Don't forget that you can submit jobs with command line arguments:
Bj.submit "./jobs/a.rb 1 foobar --force"
and that you can do powerful things by passing stdin to a job that powers
through a list of work. For instance, assume a "./jobs/bulkmail" job
resembling
STDIN.each do |line|
address = line.strip
mail_message_to address
end
then you could
stdin = [
]
Bj.submit "./script/runner ./jobs/bulkmail", :stdin => stdin
and all those emails would be sent in the background.
Bj's power is putting jobs in the background in a simple and robust fashion.
It's your task to build intelligent jobs that leverage batch processing, and
other, possibilities. The upshot of building tasks this way is that they are
quite easy to test before submitting them from inside your application.
________________________________
Install
--------------------------------
Bj can be installed two ways: as a plugin or via rubygems
plugin:
1) ./script/plugin install http://codeforpeople.rubyforge.org/svn/rails/plugins/bj
2) ./script/bj setup
gem:
1) $sudo gem install bj
2) add "require 'bj'" to config/environment.rb
3) bj setup
________________________________
Api
--------------------------------
submit jobs for background processing. 'jobs' can be a string or array of
strings. options are applied to each job in the 'jobs', and the list of
submitted jobs is always returned. options (string or symbol) can be
:rails_env => production|development|key_in_database_yml
when given this keyword causes bj to submit jobs to the
specified database. default is RAILS_ENV.
:priority => any number, including negative ones. default is zero.
:tag => a tag added to the job. simply makes searching easier.
:env => a hash specifying any additional environment vars the background
process should have.
:stdin => any stdin the background process should have. must respond_to
to_s
eg:
jobs = Bj.submit 'echo foobar', :tag => 'simple job'
jobs = Bj.submit '/bin/cat', :stdin => 'in the hat', :priority => 42
jobs = Bj.submit './script/runner ./scripts/a.rb', :rails_env => 'production'
jobs = Bj.submit './script/runner /dev/stdin',
:stdin => 'p RAILS_ENV',
:tag => 'dynamic ruby code'
jobs Bj.submit array_of_commands, :priority => 451
when jobs are run, they are run in RAILS_ROOT. various attributes are
available *only* once the job has finished. you can check whether or not a
job is finished by using the #finished method, which simple does a reload and
checks to see if the exit_status is non-nil.
eg:
jobs = Bj.submit list_of_jobs, :tag => 'important'
...
jobs.each do |job|
if job.finished?
p job.exit_status
p job.stdout
p job.stderr
end
end
See lib/bj/api.rb for more details.
________________________________
Sponsors
--------------------------------
http://quintess.com/
http://www.engineyard.com/
http://igicom.com/
http://eparklabs.com/
http://your_company.com/ <<-- (targeted marketing aimed at *you*)
________________________________
Version
--------------------------------
1.0.1
PARAMETERS
--rails_root=rails_root, -R (0 ~> rails_root=)
the rails_root will be guessed unless you set this
--rails_env=rails_env, -E (0 ~> rails_env=development)
set the rails_env
--log=log, -l (0 ~> log=STDERR)
set the logfile
--help, -h
AUTHOR
URIS
http://codeforpeople.com/lib/ruby/
http://rubyforge.org/projects/codeforpeople/
http://codeforpeople.rubyforge.org/svn/rails/plugins/