-
Notifications
You must be signed in to change notification settings - Fork 18
/
Copy pathnode_querier.rb
656 lines (584 loc) · 23.6 KB
/
node_querier.rb
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
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
require 'net/sftp'
module MotherBrain
class NodeQuerier
DISABLED_RUN_LIST_ENTRY = "recipe[disabled]".freeze
class << self
# @raise [Celluloid::DeadActorError] if Node Querier has not been started
#
# @return [Celluloid::Actor(NodeQuerier)]
def instance
MB::Application[:node_querier] or raise Celluloid::DeadActorError, "node querier not running"
end
end
extend Forwardable
include Celluloid
include MB::Logging
include MB::Mixin::Locks
include MB::Mixin::Services
finalizer :finalize_callback
def initialize
log.debug { "Node Querier starting..." }
end
# List all of the nodes on the target Chef Server
#
# @return [Array<Hash>]
def list
chef_connection.node.all
end
# Run Chef on a group of nodes, and update a job status with the result
# @param [Job] job
# @param [Array(Ridley::NodeResource)] nodes
# The collection of nodes to run Chef on
# @param [Array<String>] override_recipes
# An array of run list entries that will override the node's current run list
#
# @raise [RemoteCommandError]
def bulk_chef_run(job, nodes, override_recipes = nil)
job.set_status("Performing a chef client run on #{nodes.collect(&:name).join(', ')}")
node_successes_count = 0
node_successes = Array.new
node_failures_count = 0
node_failures = Array.new
futures = nodes.map { |node| node_querier.future(:chef_run, node.public_hostname, override_recipes: override_recipes, connector: connector_for_os(node.chef_attributes.os)) }
futures.each do |future|
begin
response = future.value
node_successes_count += 1
node_successes << response.host
rescue RemoteCommandError => error
node_failures_count += 1
node_failures << error.host
end
end
if node_failures_count > 0
abort RemoteCommandError.new("chef client run failed on #{node_failures_count} node(s) - #{node_failures.join(', ')}")
else
job.set_status("Finished chef client run on #{node_successes_count} node(s) - #{node_successes.join(', ')}")
end
end
# Return the Chef node_name of the target host. A nil value is returned if a
# node_name cannot be determined
#
# @param [String] host
# hostname of the target node
# @option options [String] :user
# a shell user that will login to each node and perform the bootstrap command on (required)
# @option options [String] :password
# the password for the shell user that will perform the bootstrap
# @option options [Array, String] :keys
# an array of keys (or a single key) to authenticate the ssh user with instead of a password
# @option options [Float] :timeout (10.0)
# timeout value for SSH bootstrap
# @option options [Boolean] :sudo (true)
# bootstrap with sudo
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @return [String, nil]
def node_name(host, options = {})
ruby_script('node_name', host, options).split("\n").last
rescue MB::RemoteScriptError
# TODO: catch auth error?
nil
end
# Run Chef-Client on the target host
#
# @param [String] host
#
# @option options [String] :user
# a shell user that will login to each node and perform the bootstrap command on (required)
# @option options [String] :password
# the password for the shell user that will perform the bootstrap
# @option options [Array, String] :keys
# an array of keys (or a single key) to authenticate the ssh user with instead of a password
# @option options [Float] :timeout (10.0)
# timeout value for SSH bootstrap
# @option options [Boolean] :sudo
# bootstrap with sudo
# @option options [String] :override_recipe
# a recipe that will override the nodes current run list
# @option options [Ridley::NodeObject] :node
# the actual node object
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @raise [RemoteCommandError] if an execution error occurs in the remote command
# @raise [RemoteCommandError] if given a blank or nil hostname
#
# @return [Ridley::HostConnector::Response]
def chef_run(host, options = {})
options = options.dup
unless host.present?
abort RemoteCommandError.new("cannot execute a chef-run without a hostname or ipaddress")
end
response = if options[:override_recipes]
override_recipes = options[:override_recipes]
cmd_recipe_syntax = override_recipes.join(',') { |recipe| "recipe[#{recipe}]" }
log.info { "Running Chef client with override runlist '#{cmd_recipe_syntax}' on: #{host}" }
chef_run_response = safe_remote(host) { chef_connection.node.execute_command(host, "chef-client --override-runlist #{cmd_recipe_syntax}", connector: options[:connector]) }
chef_run_response
else
log.info { "Running Chef client on: #{host}" }
safe_remote(host) { chef_connection.node.chef_run(host, connector: options[:connector]) }
end
if response.error?
log.info { "Failed Chef client run on: #{host} - #{response.stderr.chomp}" }
abort RemoteCommandError.new(response.stderr.chomp, host)
end
log.info { "Completed Chef client run on: #{host}" }
response
rescue Ridley::Errors::HostConnectionError => ex
log.info { "Failed Chef client run on: #{host}" }
abort RemoteCommandError.new(ex, host)
end
# Place an encrypted data bag secret on the target host
#
# @param [String] host
#
# @option options [String] :secret
# the encrypted data bag secret of the node querier's chef conn will be used
# as the default key
# @option options [String] :user
# a shell user that will login to each node and perform the bootstrap command on (required)
# @option options [String] :password
# the password for the shell user that will perform the bootstrap
# @option options [Array, String] :keys
# an array of keys (or a single key) to authenticate the ssh user with instead of a password
# @option options [Float] :timeout (10.0)
# timeout value for SSH bootstrap
# @option options [Boolean] :sudo
# bootstrap with sudo
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @raise [RemoteFileCopyError]
#
# @return [Ridley::HostConnector::Response]
def put_secret(host, options = {})
options = options.reverse_merge(secret: Application.config.chef.encrypted_data_bag_secret_path)
if options[:secret].nil? || !File.exists?(options[:secret])
return nil
end
unless host.present?
abort RemoteCommandError.new("cannot put_secret without a hostname or ipaddress")
end
response = safe_remote(host) { chef_connection.node.put_secret(host, connector: options[:connector]) }
if response.error?
log.info { "Failed to put secret file on: #{host}" }
return nil
end
log.info { "Successfully put secret file on: #{host}" }
response
end
# Executes the given command on the host using the best worker
# available for the host.
#
# @param [String] host
# @param [String] command
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @return [Ridley::HostConnection::Response]
def execute_command(host, command, options = {})
unless host.present?
abort RemoteCommandError.new("cannot execute command without a hostname or ipaddress")
end
response = safe_remote(host) { chef_connection.node.execute_command(host, command, connector: options[:connector]) }
if response.error?
log.info { "Failed to execute command on: #{host}" }
abort RemoteCommandError.new(response.stderr.chomp)
end
log.info { "Successfully executed command on: #{host}" }
response
end
# Check if the target host is registered with the Chef server. If the node does not have Chef and
# ruby installed by omnibus it will be considered unregistered.
#
# @example showing a node who is registered to Chef
# node_querier.registered?("192.168.1.101") #=> true
# @example showing a node who does not have ruby or is not registered to Chef
# node_querier.registered?("192.168.1.102") #=> false
#
# @param [String] host
# public hostname of the target node
#
# @return [Boolean]
def registered?(host)
!!registered_as(host)
end
# Returns the client name the target node is registered to Chef with.
#
# If the node does not have a client registered with the Chef server or if Chef and ruby were not installed
# by omnibus this function will return nil.
#
# @example showing a node who is registered to Chef
# node_querier.registered_as("192.168.1.101") #=> "reset.riotgames.com"
# @example showing a node who does not have ruby or is not registered to Chef
# node_querier.registered_as("192.168.1.102") #=> nil
#
# @param [String] host
# public hostname of the target node
#
# @return [String, nil]
def registered_as(host)
if (client_id = node_name(host)).nil?
return nil
end
safe_remote(host) { chef_connection.client.find(client_id).try(:name) }
end
# Asynchronously remove Chef from a target host and purge it's client and node object from the
# Chef server.
#
# @param [String] host
# public hostname of the target node
#
# @option options [Boolean] :skip_chef (false)
# skip removal of the Chef package and the contents of the installation
# directory. Setting this to true will only remove any data and configurations
# generated by running Chef client.
#
# @return [MB::JobTicket]
def async_purge(host, options = {})
job = Job.new(:purge_node)
async(:purge, job, host, options)
job.ticket
end
# Asynchronously upgrade the Omnibus installation of Chef on the given nodes
# to a specific version.
#
# @param [String] version
# the version of Chef to upgrade to
# @param [Array<Ridley::NodeObject>] nodes
# the node(s) to upgrade omnibus on
#
# @option options [Boolean] :prerelease
# whether or not to use a prerelease version of Chef
# @option options [String] :direct_url
# a URL pointing directly to a Chef package to install
#
# @return [Mb::JobTicket]
def async_upgrade_omnibus(version, nodes, options = {})
job = Job.new(:upgrade_omnibus)
async(:upgrade_omnibus, job, version, nodes, options)
job.ticket
end
# Asynchronously disable a node to stop services @host and prevent
# chef-client from being run on @host until @host is reenabled
#
# @param [String] host
# public hostname of the target node
# @param [Hash] options
#
# @option options [Boolean] :force (false) Ignore environment lock and execute anyway.
#
# @return [MB::JobTicket]
def async_disable(host, options = {})
job = Job.new(:disable_node)
async(:disable, job, host, options)
job.ticket
end
# Asynchronously enable a node
#
# @param [String] host
# public hostname of the target node
# @param [Hash] options
#
# @option options [Boolean] :force (false) Ignore environment lock and execute anyway.
#
# @return [MB::JobTicket]
def async_enable(host, options = {})
job = Job.new(:enable_node)
async(:enable, job, host, options)
job.ticket
end
# Remove Chef from a target host and purge it's client and node object from the Chef
# server.
#
# @param [MB::Job] job
# @param [String] host
# public hostname of the target node
#
# @option options [Boolean] :skip_chef (false)
# skip removal of the Chef package and the contents of the installation
# directory. Setting this to true will only remove any data and configurations
# generated by running Chef client.
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @return [MB::JobTicket]
def purge(job, host, options = {})
options = options.reverse_merge(skip_chef: false)
futures = Array.new
job.report_running("Discovering host's registered node name")
if node_name = registered_as(host)
job.set_status("Host registered as #{node_name}. Destroying client and node objects.")
futures << chef_connection.client.future(:delete, node_name)
futures << chef_connection.node.future(:delete, node_name)
else
job.set_status "Could not discover the host's node name. The host may not be registered with Chef or the " +
"embedded Ruby used to identify the node name may not be available."
end
job.set_status("Cleaning up the host's file system.")
futures << chef_connection.node.future(:uninstall_chef, host, options.slice(:skip_chef, :connector))
begin
safe_remote(host) { futures.map(&:value) }
rescue RemoteCommandError => e
job.report_failure
end
job.report_success
ensure
job.terminate if job && job.alive?
end
# Upgrades the Omnibus installation of Chef on a specific node(s) to
# a specific version.
#
# @param [MB::Job] job
# @param [String] version
# the version of Chef to upgrade to
# @param [Array<Ridley::NodeObject>] nodes
# the node(s) to upgrade omnibus on
#
# @option options [Boolean] :prerelease
# whether or not to use a prerelease version of Chef
# @option options [String] :direct_url
# a URL pointing directly to a Chef package to install
#
# @return [MB::Job]
def upgrade_omnibus(job, version, nodes, options = {})
futures = Array.new
options = options.merge(chef_version: version)
hostnames = nodes.collect(&:public_hostname)
job.report_running("Upgrading Omnibus Chef installation on #{hostnames}")
hostnames.each do |hostname|
futures << chef_connection.node.future(:update_omnibus, hostname, options)
end
begin
safe_remote { futures.map(&:value) }
rescue RemoteCommandError => e
job.report_failure
end
job.report_success
ensure
job.terminate if job && job.alive?
end
# Remove explicit service state on @host and remove disabled entry
# from run list to allow chef-client to run on @host
#
# @param [MB::Job] job
# @param [String] host
# public hostname of the target node
# @param [Hash] options
#
# @option options [Boolean] :force (false) Ignore environment lock and execute anyway.
#
def enable(job, host, options = {})
job.report_running("Discovering host's registered node name")
node_name = registered_as(host)
if !node_name
# TODO auth could fail and cause this to throw
job.report_failure("Could not discover the host's node name. The host may not be " +
"registered with Chef or the embedded Ruby used to identify the " +
"node name may not be available. #{host} was not enabled!")
end
job.set_status("Host registered as #{node_name}.")
node = fetch_node(job, node_name)
required_run_list = []
success = false
chef_synchronize(chef_environment: node.chef_environment, force: options[:force], job: job) do
if node.run_list.include?(DISABLED_RUN_LIST_ENTRY)
required_run_list = on_dynamic_services(job, node) do |dynamic_service, plugin|
dynamic_service.remove_node_state_change(job,
plugin,
node,
false)
end
if !required_run_list.empty?
self.bulk_chef_run(job, [node], required_run_list.flatten.uniq)
end
node.run_list = node.run_list.reject { |r| r == DISABLED_RUN_LIST_ENTRY }
if node.save
job.set_status "#{node.name} enabled successfully."
success = true
else
job.set_status "#{node.name} did not save! Disabled run_list entry was unable to be removed to the node."
end
else
job.set_status("#{node.name} is not disabled. No need to enable.")
success = true
end
end
job.report_boolean(success)
rescue MotherBrain::ResourceLocked => e
job.report_failure e.message
ensure
job.terminate if job && job.alive?
end
# Stop services on @host and prevent chef-client from being run on
# @host until @host is reenabled
#
# @param [MB::Job] job
# @param [String] host
# public hostname of the target node
# @param [Hash] options
#
# @option options [Boolean] :force (false) Ignore environment lock and execute anyway.
#
def disable(job, host, options = {})
job.report_running("Discovering host's registered node name")
node_name = registered_as(host)
if !node_name
# TODO auth could fail and cause this to throw
job.report_failure("Could not discover the host's node name. The host may not be " +
"registered with Chef or the embedded Ruby used to identify the " +
"node name may not be available. #{host} was not disabled!")
end
job.set_status("Host registered as #{node_name}.")
node = fetch_node(job, node_name)
required_run_list = []
success = false
chef_synchronize(chef_environment: node.chef_environment, force: options[:force], job: job) do
if node.run_list.include?(DISABLED_RUN_LIST_ENTRY)
job.set_status("#{node.name} is already disabled.")
success = true
else
required_run_list = on_dynamic_services(job, node) do |dynamic_service, plugin|
dynamic_service.node_state_change(job,
plugin,
node,
MB::Gear::DynamicService::STOP,
false)
end
end
if !success
if !required_run_list.empty?
job.set_status "Running chef with the following run list: #{required_run_list.inspect}"
self.bulk_chef_run(job, [node], required_run_list)
else
job.set_status "No recipes required to run."
end
node.run_list = [DISABLED_RUN_LIST_ENTRY].concat(node.run_list)
if node.save
job.set_status "#{node.name} disabled."
success = true
else
job.set_status "#{node.name} did not save! Disabled run_list entry was unable to be added to the node."
end
end
end
job.report_boolean(success)
rescue MotherBrain::ResourceLocked => e
job.report_failure e.message
ensure
job.terminate if job && job.alive?
end
# Returns a String representing the best connector
# type to use when communicating with a given node
#
# @param os [String]
# the operating system
#
# @return [String]
def connector_for_os(os)
case os
when "windows"
Ridley::HostCommander::DEFAULT_WINDOWS_CONNECTOR
when "linux"
Ridley::HostCommander::DEFAULT_LINUX_CONNECTOR
else
nil
end
end
private
def finalize_callback
log.debug { "Node Querier stopping..." }
end
# Run a Ruby script on the target host and return the result of STDOUT. Only scripts
# that are located in the Mother Brain scripts directory can be used and they should
# be identified just by their filename minus the extension
#
# @example
# node_querier.ruby_script('node_name', '33.33.33.10') => 'vagrant.localhost'
#
# @param [String] name
# name of the script to run on the target node
# @param [String] host
# hostname of the target node
# the MotherBrain scripts directory
# @option options [String] :user
# a shell user that will login to each node and perform the bootstrap command on (required)
# @option options [String] :password
# the password for the shell user that will perform the bootstrap
# @option options [Array, String] :keys
# an array of keys (or a single key) to authenticate the ssh user with instead of a password
# @option options [Float] :timeout (10.0)
# timeout value for SSH bootstrap
# @option options [Boolean] :sudo (true)
# bootstrap with sudo
# @option options [String] :connector
# a connector type for the chef connection to prefer
#
# @raise [RemoteScriptError] if there was an error in execution
# @raise [RuntimeError] if an unknown response is returned from Ridley
#
# @note
# Use {#_ruby_script_} if the caller of this function is same actor as the receiver. You will
# not be able to rescue from the RemoteScriptError thrown by {#ruby_script} but you will
# be able to rescue from {#_ruby_script_}.
#
# @return [String]
def ruby_script(name, host, options = {})
name = name.split('.rb')[0]
lines = File.readlines(MB.scripts.join("#{name}.rb"))
command_lines = lines.collect { |line| line.gsub('"', "'").strip.chomp }
unless host.present?
abort RemoteCommandError.new("cannot execute a ruby_script without a hostname or ipaddress")
end
response = safe_remote(host) { chef_connection.node.ruby_script(host, command_lines, connector: options[:connector]) }
if response.error?
raise RemoteScriptError.new(response.stderr.chomp)
end
response.stdout.chomp
end
def on_dynamic_services(job, node)
[].tap do |required_run_list|
node.run_list.each do |run_list_entry|
next if run_list_entry == DISABLED_RUN_LIST_ENTRY
plugin = plugin_manager.for_run_list_entry(run_list_entry, node.chef_environment, remote: true)
if plugin.nil?
job.report_failure("Could not find plugin for #{run_list_entry}. Aborting command.")
end
plugin.components.each do |component|
services = component.gears(MB::Gear::Service)
services.each do |service|
if service.dynamic_service?
if component.group(service.service_group).includes_recipe? run_list_entry
dynamic_service = service.to_dynamic_service
yield(dynamic_service, plugin)
required_run_list << service.service_recipe
end
else
job.report_failure("Service #{component.name}.#{service.name} is not a dynamic service;" +
" MotherBrain cannot set the state of this service automatically." +
" Aborting command.")
end
end
end
end
end.flatten.uniq
end
def fetch_node(job, node_name)
node = safe_remote(node_name) { chef_connection.node.find(node_name) }
rescue RemoteCommandError => e
job.report_failure("Encountered error retrieving the node object.")
end
def safe_remote(host = nil)
yield
rescue Exception => e
msg = "Unhandled Exception: [#{e.class}] #{e.message}"
msg = "[#{host}] #{msg}" if host
log.warn { msg }
log.debug { e.backtrace.join("\n") }
abort RemoteCommandError.new(msg, host)
end
end
end