sensu-plugins-mongodb 0.0.1.alpha.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml ADDED
@@ -0,0 +1,7 @@
1
+ ---
2
+ SHA1:
3
+ metadata.gz: 9e030b5cd0b60de464db1e487012786502b38f93
4
+ data.tar.gz: 97172991de37fa87c9fb7c88ff2c07211e4c842f
5
+ SHA512:
6
+ metadata.gz: fd2e39b1fd3a0b6f7381a97ca830ba7621a05806aa768b280edce48aaf53c78b6567cd330f86298957f45999940247e8e35a8af7a4a2a08f2d149d53f89c815d
7
+ data.tar.gz: 5cc8bdf01bd2de21b28861c1f36516f9919f70cb7b73281bfc504a83ac35089740f6595895a70a5298aa2f51e22264386fc23dce5fd4e7214d22354b179cdea4
checksums.yaml.gz.sig ADDED
Binary file
data.tar.gz.sig ADDED
Binary file
data/CHANGELOG.md ADDED
@@ -0,0 +1,11 @@
1
+ #### 0.0.1.alpha.1
2
+
3
+ * identical code to the community-plugins repo
4
+
5
+ #### 0.0.1.alpha.2
6
+
7
+ * updated Vagrantfile
8
+ * add metadata to gem
9
+ * update Readme
10
+ * add new version and bump tasks
11
+ * add new version module
data/LICENSE ADDED
@@ -0,0 +1,22 @@
1
+ Copyright (c) 2015 devops@yieldbot.com
2
+
3
+ MIT License
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining
6
+ a copy of this software and associated documentation files (the
7
+ "Software"), to deal in the Software without restriction, including
8
+ without limitation the rights to use, copy, modify, merge, publish,
9
+ distribute, sublicense, and/or sell copies of the Software, and to
10
+ permit persons to whom the Software is furnished to do so, subject to
11
+ the following conditions:
12
+
13
+ The above copyright notice and this permission notice shall be
14
+ included in all copies or substantial portions of the Software.
15
+
16
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
17
+ EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
18
+ MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
19
+ NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
20
+ LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
21
+ OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
22
+ WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
data/README.md ADDED
@@ -0,0 +1,54 @@
1
+ ## Sensu-Plugins-mongodb
2
+
3
+ [![Build Status](https://travis-ci.org/sensu-plugins/sensu-plugins-mongodb.svg?branch=master)](https://travis-ci.org/sensu-plugins/sensu-plugins-mongodb)
4
+ [![Gem Version](https://badge.fury.io/rb/sensu-plugins-mongodb.svg)](http://badge.fury.io/rb/sensu-plugins-mongodb)
5
+ [![Code Climate](https://codeclimate.com/github/sensu-plugins/sensu-plugins-mongodb/badges/gpa.svg)](https://codeclimate.com/github/sensu-plugins/sensu-plugins-mongodb)
6
+ [![Test Coverage](https://codeclimate.com/github/sensu-plugins/sensu-plugins-mongodb/badges/coverage.svg)](https://codeclimate.com/github/sensu-plugins/sensu-plugins-mongodb)
7
+ [![Dependency Status](https://gemnasium.com/sensu-plugins/sensu-plugins-mongodb.svg)](https://gemnasium.com/sensu-plugins/sensu-plugins-mongodb)
8
+
9
+ ## Functionality
10
+
11
+ ## Files
12
+ * bin/check-mongodb.py
13
+ * bin/metrics-mongodb.rb
14
+
15
+ ## Usage
16
+
17
+ ## Installation
18
+
19
+ Add the public key (if you haven’t already) as a trusted certificate
20
+
21
+ ```
22
+ gem cert --add <(curl -Ls https://raw.githubusercontent.com/sensu-plugins/sensu-plugins.github.io/master/certs/sensu-plugins.pem)
23
+ gem install sensu-plugins-mongodb -P MediumSecurity
24
+ ```
25
+
26
+ You can also download the key from /certs/ within each repository.
27
+
28
+ #### Rubygems
29
+
30
+ `gem install sensu-plugins-mongodb`
31
+
32
+ #### Bundler
33
+
34
+ Add *sensu-plugins-mongodb* to your Gemfile and run `bundle install` or `bundle update`
35
+
36
+ #### Chef
37
+
38
+ Using the Sensu **sensu_gem** LWRP
39
+ ```
40
+ sensu_gem 'sensu-plugins-mongodb' do
41
+ options('--prerelease')
42
+ version '0.0.1.alpha.1'
43
+ end
44
+ ```
45
+
46
+ Using the Chef **gem_package** resource
47
+ ```
48
+ gem_package 'sensu-plugins-mongodb' do
49
+ options('--prerelease')
50
+ version '0.0.1.alpha.1'
51
+ end
52
+ ```
53
+
54
+ ## Notes
@@ -0,0 +1,1449 @@
1
+ #!/usr/bin/env python
2
+
3
+ #
4
+ # A MongoDB Nagios check script
5
+ #
6
+
7
+ # Script idea taken from a Tag1 script I found and I modified it a lot
8
+ #
9
+ # Main Author
10
+ # - Mike Zupan <mike@zcentric.com>
11
+ # Contributers
12
+ # - Frank Brandewiede <brande@travel-iq.com> <brande@bfiw.de> <brande@novolab.de>
13
+ # - Sam Perman <sam@brightcove.com>
14
+ # - Shlomo Priymak <shlomoid@gmail.com>
15
+ # - @jhoff909 on github
16
+ # - @jbraeuer on github
17
+ # - Dag Stockstad <dag.stockstad@gmail.com>
18
+ # - @Andor on github
19
+ # - Steven Richards - Captainkrtek on Github <sbrichards@mit.edu>
20
+ #
21
+
22
+ # License: BSD
23
+ # Copyright (c) 2012, Mike Zupan <mike@zcentric.com>
24
+ # All rights reserved.
25
+ # Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:
26
+ #
27
+ # Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.
28
+ # Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the
29
+ # documentation and/or other materials provided with the distribution.
30
+ # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
31
+ # THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS
32
+ # BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
33
+ # GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
34
+ # STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
35
+ #
36
+ # README: https://github.com/mzupan/nagios-plugin-mongodb/blob/master/LICENSE
37
+
38
+ # #RED
39
+ import sys
40
+ import time
41
+ import optparse
42
+ import textwrap
43
+ import re
44
+ import os
45
+
46
+ try:
47
+ import pymongo
48
+ except ImportError, e:
49
+ print e
50
+ sys.exit(2)
51
+
52
+ # As of pymongo v 1.9 the SON API is part of the BSON package, therefore attempt
53
+ # to import from there and fall back to pymongo in cases of older pymongo
54
+ if pymongo.version >= "1.9":
55
+ import bson.son as son
56
+ else:
57
+ import pymongo.son as son
58
+
59
+
60
+ #
61
+ # thanks to http://stackoverflow.com/a/1229667/72987
62
+ #
63
+ def optional_arg(arg_default):
64
+ def func(option, opt_str, value, parser):
65
+ if parser.rargs and not parser.rargs[0].startswith('-'):
66
+ val = parser.rargs[0]
67
+ parser.rargs.pop(0)
68
+ else:
69
+ val = arg_default
70
+ setattr(parser.values, option.dest, val)
71
+ return func
72
+
73
+
74
+ def performance_data(perf_data, params):
75
+ data = ''
76
+ if perf_data:
77
+ data = " |"
78
+ for p in params:
79
+ p += (None, None, None, None)
80
+ param, param_name, warning, critical = p[0:4]
81
+ data += "%s=%s" % (param_name, str(param))
82
+ if warning or critical:
83
+ warning = warning or 0
84
+ critical = critical or 0
85
+ data += ";%s;%s" % (warning, critical)
86
+
87
+ data += " "
88
+
89
+ return data
90
+
91
+
92
+ def numeric_type(param):
93
+ if ((type(param) == float or type(param) == int or param == None)):
94
+ return True
95
+ return False
96
+
97
+
98
+ def check_levels(param, warning, critical, message, ok=[]):
99
+ if (numeric_type(critical) and numeric_type(warning)):
100
+ if param >= critical:
101
+ print "CRITICAL - " + message
102
+ sys.exit(2)
103
+ elif param >= warning:
104
+ print "WARNING - " + message
105
+ sys.exit(1)
106
+ else:
107
+ print "OK - " + message
108
+ sys.exit(0)
109
+ else:
110
+ if param in critical:
111
+ print "CRITICAL - " + message
112
+ sys.exit(2)
113
+
114
+ if param in warning:
115
+ print "WARNING - " + message
116
+ sys.exit(1)
117
+
118
+ if param in ok:
119
+ print "OK - " + message
120
+ sys.exit(0)
121
+
122
+ # unexpected param value
123
+ print "CRITICAL - Unexpected value : %d" % param + "; " + message
124
+ return 2
125
+
126
+
127
+ def get_server_status(con):
128
+ try:
129
+ set_read_preference(con.admin)
130
+ data = con.admin.command(pymongo.son_manipulator.SON([('serverStatus', 1)]))
131
+ except:
132
+ data = con.admin.command(son.SON([('serverStatus', 1)]))
133
+ return data
134
+
135
+
136
+ def main(argv):
137
+ p = optparse.OptionParser(conflict_handler="resolve", description="This Nagios plugin checks the health of mongodb.")
138
+
139
+ p.add_option('-H', '--host', action='store', type='string', dest='host', default='127.0.0.1', help='The hostname you want to connect to')
140
+ p.add_option('-P', '--port', action='store', type='int', dest='port', default=27017, help='The port mongodb is runnung on')
141
+ p.add_option('-u', '--user', action='store', type='string', dest='user', default=None, help='The username you want to login as')
142
+ p.add_option('-p', '--pass', action='store', type='string', dest='passwd', default=None, help='The password you want to use for that user')
143
+ p.add_option('-W', '--warning', action='store', dest='warning', default=None, help='The warning threshold we want to set')
144
+ p.add_option('-C', '--critical', action='store', dest='critical', default=None, help='The critical threshold we want to set')
145
+ p.add_option('-A', '--action', action='store', type='choice', dest='action', default='connect', help='The action you want to take',
146
+ choices=['connect', 'connections', 'replication_lag', 'replication_lag_percent', 'replset_state', 'memory', 'memory_mapped', 'lock',
147
+ 'flushing', 'last_flush_time', 'index_miss_ratio', 'databases', 'collections', 'database_size', 'database_indexes', 'collection_indexes', 'collection_size',
148
+ 'queues', 'oplog', 'journal_commits_in_wl', 'write_data_files', 'journaled', 'opcounters', 'current_lock', 'replica_primary', 'page_faults',
149
+ 'asserts', 'queries_per_second', 'page_faults', 'chunks_balance', 'connect_primary', 'collection_state', 'row_count', 'replset_quorum'])
150
+ p.add_option('--max-lag', action='store_true', dest='max_lag', default=False, help='Get max replication lag (for replication_lag action only)')
151
+ p.add_option('--mapped-memory', action='store_true', dest='mapped_memory', default=False, help='Get mapped memory instead of resident (if resident memory can not be read)')
152
+ p.add_option('-D', '--perf-data', action='store_true', dest='perf_data', default=False, help='Enable output of Nagios performance data')
153
+ p.add_option('-d', '--database', action='store', dest='database', default='admin', help='Specify the database to check')
154
+ p.add_option('--all-databases', action='store_true', dest='all_databases', default=False, help='Check all databases (action database_size)')
155
+ p.add_option('-s', '--ssl', dest='ssl', default=False, action='callback', callback=optional_arg(True), help='Connect using SSL')
156
+ p.add_option('-r', '--replicaset', dest='replicaset', default=None, action='callback', callback=optional_arg(True), help='Connect to replicaset')
157
+ p.add_option('-q', '--querytype', action='store', dest='query_type', default='query', help='The query type to check [query|insert|update|delete|getmore|command] from queries_per_second')
158
+ p.add_option('-c', '--collection', action='store', dest='collection', default='admin', help='Specify the collection to check')
159
+ p.add_option('-T', '--time', action='store', type='int', dest='sample_time', default=1, help='Time used to sample number of pages faults')
160
+
161
+ options, arguments = p.parse_args()
162
+ host = options.host
163
+ port = options.port
164
+ user = options.user
165
+ passwd = options.passwd
166
+ query_type = options.query_type
167
+ collection = options.collection
168
+ sample_time = options.sample_time
169
+ if (options.action == 'replset_state'):
170
+ warning = str(options.warning or "")
171
+ critical = str(options.critical or "")
172
+ else:
173
+ warning = float(options.warning or 0)
174
+ critical = float(options.critical or 0)
175
+
176
+ action = options.action
177
+ perf_data = options.perf_data
178
+ max_lag = options.max_lag
179
+ database = options.database
180
+ ssl = options.ssl
181
+ replicaset = options.replicaset
182
+
183
+ if action == 'replica_primary' and replicaset is None:
184
+ return "replicaset must be passed in when using replica_primary check"
185
+ elif not action == 'replica_primary' and replicaset:
186
+ return "passing a replicaset while not checking replica_primary does not work"
187
+
188
+ #
189
+ # moving the login up here and passing in the connection
190
+ #
191
+ start = time.time()
192
+ err, con = mongo_connect(host, port, ssl, user, passwd, replicaset)
193
+ if err != 0:
194
+ return err
195
+
196
+ conn_time = time.time() - start
197
+ conn_time = round(conn_time, 0)
198
+
199
+ if action == "connections":
200
+ return check_connections(con, warning, critical, perf_data)
201
+ elif action == "replication_lag":
202
+ return check_rep_lag(con, host, port, warning, critical, False, perf_data, max_lag, user, passwd)
203
+ elif action == "replication_lag_percent":
204
+ return check_rep_lag(con, host, port, warning, critical, True, perf_data, max_lag, user, passwd)
205
+ elif action == "replset_state":
206
+ return check_replset_state(con, perf_data, warning, critical)
207
+ elif action == "memory":
208
+ return check_memory(con, warning, critical, perf_data, options.mapped_memory)
209
+ elif action == "memory_mapped":
210
+ return check_memory_mapped(con, warning, critical, perf_data)
211
+ elif action == "queues":
212
+ return check_queues(con, warning, critical, perf_data)
213
+ elif action == "lock":
214
+ return check_lock(con, warning, critical, perf_data)
215
+ elif action == "current_lock":
216
+ return check_current_lock(con, host, warning, critical, perf_data)
217
+ elif action == "flushing":
218
+ return check_flushing(con, warning, critical, True, perf_data)
219
+ elif action == "last_flush_time":
220
+ return check_flushing(con, warning, critical, False, perf_data)
221
+ elif action == "index_miss_ratio":
222
+ index_miss_ratio(con, warning, critical, perf_data)
223
+ elif action == "databases":
224
+ return check_databases(con, warning, critical, perf_data)
225
+ elif action == "collections":
226
+ return check_collections(con, warning, critical, perf_data)
227
+ elif action == "oplog":
228
+ return check_oplog(con, warning, critical, perf_data)
229
+ elif action == "journal_commits_in_wl":
230
+ return check_journal_commits_in_wl(con, warning, critical, perf_data)
231
+ elif action == "database_size":
232
+ if options.all_databases:
233
+ return check_all_databases_size(con, warning, critical, perf_data)
234
+ else:
235
+ return check_database_size(con, database, warning, critical, perf_data)
236
+ elif action == "database_indexes":
237
+ return check_database_indexes(con, database, warning, critical, perf_data)
238
+ elif action == "collection_indexes":
239
+ return check_collection_indexes(con, database, collection, warning, critical, perf_data)
240
+ elif action == "collection_size":
241
+ return check_collection_size(con, database, collection, warning, critical, perf_data)
242
+ elif action == "journaled":
243
+ return check_journaled(con, warning, critical, perf_data)
244
+ elif action == "write_data_files":
245
+ return check_write_to_datafiles(con, warning, critical, perf_data)
246
+ elif action == "opcounters":
247
+ return check_opcounters(con, host, warning, critical, perf_data)
248
+ elif action == "asserts":
249
+ return check_asserts(con, host, warning, critical, perf_data)
250
+ elif action == "replica_primary":
251
+ return check_replica_primary(con, host, warning, critical, perf_data, replicaset)
252
+ elif action == "queries_per_second":
253
+ return check_queries_per_second(con, query_type, warning, critical, perf_data)
254
+ elif action == "page_faults":
255
+ check_page_faults(con, sample_time, warning, critical, perf_data)
256
+ elif action == "chunks_balance":
257
+ chunks_balance(con, database, collection, warning, critical)
258
+ elif action == "connect_primary":
259
+ return check_connect_primary(con, warning, critical, perf_data)
260
+ elif action == "collection_state":
261
+ return check_collection_state(con, database, collection)
262
+ elif action == "row_count":
263
+ return check_row_count(con, database, collection, warning, critical, perf_data)
264
+ elif action == "replset_quorum":
265
+ return check_replset_quorum(con, perf_data)
266
+ else:
267
+ return check_connect(host, port, warning, critical, perf_data, user, passwd, conn_time)
268
+
269
+
270
+ def mongo_connect(host=None, port=None, ssl=False, user=None, passwd=None, replica=None):
271
+ try:
272
+ # ssl connection for pymongo > 2.3
273
+ if pymongo.version >= "2.3":
274
+ if replica is None:
275
+ con = pymongo.MongoClient(host, port)
276
+ else:
277
+ con = pymongo.Connection(host, port, read_preference=pymongo.ReadPreference.SECONDARY, ssl=ssl, replicaSet=replica, network_timeout=10)
278
+ else:
279
+ if replica is None:
280
+ con = pymongo.Connection(host, port, slave_okay=True, network_timeout=10)
281
+ else:
282
+ con = pymongo.Connection(host, port, slave_okay=True, network_timeout=10)
283
+ #con = pymongo.Connection(host, port, slave_okay=True, replicaSet=replica, network_timeout=10)
284
+
285
+ if user and passwd:
286
+ db = con["admin"]
287
+ if not db.authenticate(user, passwd):
288
+ sys.exit("Username/Password incorrect")
289
+ except Exception, e:
290
+ if isinstance(e, pymongo.errors.AutoReconnect) and str(e).find(" is an arbiter") != -1:
291
+ # We got a pymongo AutoReconnect exception that tells us we connected to an Arbiter Server
292
+ # This means: Arbiter is reachable and can answer requests/votes - this is all we need to know from an arbiter
293
+ print "OK - State: 7 (Arbiter)"
294
+ sys.exit(0)
295
+ return exit_with_general_critical(e), None
296
+ return 0, con
297
+
298
+
299
+ def exit_with_general_warning(e):
300
+ if isinstance(e, SystemExit):
301
+ return e
302
+ else:
303
+ print "WARNING - General MongoDB warning:", e
304
+ return 1
305
+
306
+
307
+ def exit_with_general_critical(e):
308
+ if isinstance(e, SystemExit):
309
+ return e
310
+ else:
311
+ print "CRITICAL - General MongoDB Error:", e
312
+ return 2
313
+
314
+
315
+ def set_read_preference(db):
316
+ if pymongo.version >= "2.1":
317
+ db.read_preference = pymongo.ReadPreference.SECONDARY
318
+
319
+
320
+ def check_connect(host, port, warning, critical, perf_data, user, passwd, conn_time):
321
+ warning = warning or 3
322
+ critical = critical or 6
323
+ message = "Connection took %i seconds" % conn_time
324
+ message += performance_data(perf_data, [(conn_time, "connection_time", warning, critical)])
325
+
326
+ return check_levels(conn_time, warning, critical, message)
327
+
328
+
329
+ def check_connections(con, warning, critical, perf_data):
330
+ warning = warning or 80
331
+ critical = critical or 95
332
+ try:
333
+ data = get_server_status(con)
334
+
335
+ current = float(data['connections']['current'])
336
+ available = float(data['connections']['available'])
337
+
338
+ used_percent = int(float(current / (available + current)) * 100)
339
+ message = "%i percent (%i of %i connections) used" % (used_percent, current, current + available)
340
+ message += performance_data(perf_data, [(used_percent, "used_percent", warning, critical),
341
+ (current, "current_connections"),
342
+ (available, "available_connections")])
343
+ return check_levels(used_percent, warning, critical, message)
344
+
345
+ except Exception, e:
346
+ return exit_with_general_critical(e)
347
+
348
+
349
+ def check_rep_lag(con, host, port, warning, critical, percent, perf_data, max_lag, user, passwd):
350
+ # Get mongo to tell us replica set member name when connecting locally
351
+ if "127.0.0.1" == host:
352
+ host = con.admin.command("ismaster","1")["me"].split(':')[0]
353
+
354
+ if percent:
355
+ warning = warning or 50
356
+ critical = critical or 75
357
+ else:
358
+ warning = warning or 600
359
+ critical = critical or 3600
360
+ rs_status = {}
361
+ slaveDelays = {}
362
+ try:
363
+ set_read_preference(con.admin)
364
+
365
+ # Get replica set status
366
+ try:
367
+ rs_status = con.admin.command("replSetGetStatus")
368
+ except pymongo.errors.OperationFailure, e:
369
+ if e.code == None and str(e).find('failed: not running with --replSet"'):
370
+ print "OK - Not running with replSet"
371
+ return 0
372
+
373
+ serverVersion = tuple(con.server_info()['version'].split('.'))
374
+ if serverVersion >= tuple("2.0.0".split(".")):
375
+ #
376
+ # check for version greater then 2.0
377
+ #
378
+ rs_conf = con.local.system.replset.find_one()
379
+ for member in rs_conf['members']:
380
+ if member.get('slaveDelay') is not None:
381
+ slaveDelays[member['host']] = member.get('slaveDelay')
382
+ else:
383
+ slaveDelays[member['host']] = 0
384
+
385
+ # Find the primary and/or the current node
386
+ primary_node = None
387
+ host_node = None
388
+
389
+ for member in rs_status["members"]:
390
+ if member["stateStr"] == "PRIMARY":
391
+ primary_node = member
392
+ if member["name"].split(':')[0] == host and int(member["name"].split(':')[1]) == port:
393
+ host_node = member
394
+
395
+ # Check if we're in the middle of an election and don't have a primary
396
+ if primary_node is None:
397
+ print "WARNING - No primary defined. In an election?"
398
+ return 1
399
+
400
+ # Check if we failed to find the current host
401
+ # below should never happen
402
+ if host_node is None:
403
+ print "CRITICAL - Unable to find host '" + host + "' in replica set."
404
+ return 2
405
+
406
+ # Is the specified host the primary?
407
+ if host_node["stateStr"] == "PRIMARY":
408
+ if max_lag == False:
409
+ print "OK - This is the primary."
410
+ return 0
411
+ else:
412
+ #get the maximal replication lag
413
+ data = ""
414
+ maximal_lag = 0
415
+ for member in rs_status['members']:
416
+ if not member['stateStr'] == "ARBITER":
417
+ lastSlaveOpTime = member['optimeDate']
418
+ replicationLag = abs(primary_node["optimeDate"] - lastSlaveOpTime).seconds - slaveDelays[member['name']]
419
+ data = data + member['name'] + " lag=%d;" % replicationLag
420
+ maximal_lag = max(maximal_lag, replicationLag)
421
+ if percent:
422
+ err, con = mongo_connect(primary_node['name'].split(':')[0], int(primary_node['name'].split(':')[1]), False, user, passwd)
423
+ if err != 0:
424
+ return err
425
+ primary_timediff = replication_get_time_diff(con)
426
+ maximal_lag = int(float(maximal_lag) / float(primary_timediff) * 100)
427
+ message = "Maximal lag is " + str(maximal_lag) + " percents"
428
+ message += performance_data(perf_data, [(maximal_lag, "replication_lag_percent", warning, critical)])
429
+ else:
430
+ message = "Maximal lag is " + str(maximal_lag) + " seconds"
431
+ message += performance_data(perf_data, [(maximal_lag, "replication_lag", warning, critical)])
432
+ return check_levels(maximal_lag, warning, critical, message)
433
+ elif host_node["stateStr"] == "ARBITER":
434
+ print "OK - This is an arbiter"
435
+ return 0
436
+
437
+ # Find the difference in optime between current node and PRIMARY
438
+
439
+ optime_lag = abs(primary_node["optimeDate"] - host_node["optimeDate"])
440
+
441
+ if host_node['name'] in slaveDelays:
442
+ slave_delay = slaveDelays[host_node['name']]
443
+ elif host_node['name'].endswith(':27017') and host_node['name'][:-len(":27017")] in slaveDelays:
444
+ slave_delay = slaveDelays[host_node['name'][:-len(":27017")]]
445
+ else:
446
+ raise Exception("Unable to determine slave delay for {0}".format(host_node['name']))
447
+
448
+ try: # work starting from python2.7
449
+ lag = optime_lag.total_seconds()
450
+ except:
451
+ lag = float(optime_lag.seconds + optime_lag.days * 24 * 3600)
452
+
453
+ if percent:
454
+ err, con = mongo_connect(primary_node['name'].split(':')[0], int(primary_node['name'].split(':')[1]), False, user, passwd)
455
+ if err != 0:
456
+ return err
457
+ primary_timediff = replication_get_time_diff(con)
458
+ if primary_timediff != 0:
459
+ lag = int(float(lag) / float(primary_timediff) * 100)
460
+ else:
461
+ lag = 0
462
+ message = "Lag is " + str(lag) + " percents"
463
+ message += performance_data(perf_data, [(lag, "replication_lag_percent", warning, critical)])
464
+ else:
465
+ message = "Lag is " + str(lag) + " seconds"
466
+ message += performance_data(perf_data, [(lag, "replication_lag", warning, critical)])
467
+ return check_levels(lag, warning + slaveDelays[host_node['name']], critical + slaveDelays[host_node['name']], message)
468
+ else:
469
+ #
470
+ # less than 2.0 check
471
+ #
472
+ # Get replica set status
473
+ rs_status = con.admin.command("replSetGetStatus")
474
+
475
+ # Find the primary and/or the current node
476
+ primary_node = None
477
+ host_node = None
478
+ for member in rs_status["members"]:
479
+ if member["stateStr"] == "PRIMARY":
480
+ primary_node = (member["name"], member["optimeDate"])
481
+ if member["name"].split(":")[0].startswith(host):
482
+ host_node = member
483
+
484
+ # Check if we're in the middle of an election and don't have a primary
485
+ if primary_node is None:
486
+ print "WARNING - No primary defined. In an election?"
487
+ sys.exit(1)
488
+
489
+ # Is the specified host the primary?
490
+ if host_node["stateStr"] == "PRIMARY":
491
+ print "OK - This is the primary."
492
+ sys.exit(0)
493
+
494
+ # Find the difference in optime between current node and PRIMARY
495
+ optime_lag = abs(primary_node[1] - host_node["optimeDate"])
496
+ lag = optime_lag.seconds
497
+ if percent:
498
+ err, con = mongo_connect(primary_node['name'].split(':')[0], int(primary_node['name'].split(':')[1]))
499
+ if err != 0:
500
+ return err
501
+ primary_timediff = replication_get_time_diff(con)
502
+ lag = int(float(lag) / float(primary_timediff) * 100)
503
+ message = "Lag is " + str(lag) + " percents"
504
+ message += performance_data(perf_data, [(lag, "replication_lag_percent", warning, critical)])
505
+ else:
506
+ message = "Lag is " + str(lag) + " seconds"
507
+ message += performance_data(perf_data, [(lag, "replication_lag", warning, critical)])
508
+ return check_levels(lag, warning, critical, message)
509
+
510
+ except Exception, e:
511
+ return exit_with_general_critical(e)
512
+
513
+
514
+ def check_memory(con, warning, critical, perf_data, mapped_memory):
515
+ #
516
+ # These thresholds are basically meaningless, and must be customized to your system's ram
517
+ #
518
+ warning = warning or 8
519
+ critical = critical or 16
520
+ try:
521
+ data = get_server_status(con)
522
+ if not data['mem']['supported'] and not mapped_memory:
523
+ print "OK - Platform not supported for memory info"
524
+ return 0
525
+ #
526
+ # convert to gigs
527
+ #
528
+ message = "Memory Usage:"
529
+ try:
530
+ mem_resident = float(data['mem']['resident']) / 1024.0
531
+ message += " %.2fGB resident," % (mem_resident)
532
+ except:
533
+ mem_resident = 0
534
+ message += " resident unsupported,"
535
+ try:
536
+ mem_virtual = float(data['mem']['virtual']) / 1024.0
537
+ message += " %.2fGB virtual," % mem_virtual
538
+ except:
539
+ mem_virtual = 0
540
+ message += " virtual unsupported,"
541
+ try:
542
+ mem_mapped = float(data['mem']['mapped']) / 1024.0
543
+ message += " %.2fGB mapped," % mem_mapped
544
+ except:
545
+ mem_mapped = 0
546
+ message += " mapped unsupported,"
547
+ try:
548
+ mem_mapped_journal = float(data['mem']['mappedWithJournal']) / 1024.0
549
+ message += " %.2fGB mappedWithJournal" % mem_mapped_journal
550
+ except:
551
+ mem_mapped_journal = 0
552
+ message += performance_data(perf_data, [("%.2f" % mem_resident, "memory_usage", warning, critical),
553
+ ("%.2f" % mem_mapped, "memory_mapped"), ("%.2f" % mem_virtual, "memory_virtual"), ("%.2f" % mem_mapped_journal, "mappedWithJournal")])
554
+ #added for unsupported systems like Solaris
555
+ if mapped_memory and mem_resident == 0:
556
+ return check_levels(mem_mapped, warning, critical, message)
557
+ else:
558
+ return check_levels(mem_resident, warning, critical, message)
559
+
560
+ except Exception, e:
561
+ return exit_with_general_critical(e)
562
+
563
+
564
+ def check_memory_mapped(con, warning, critical, perf_data):
565
+ #
566
+ # These thresholds are basically meaningless, and must be customized to your application
567
+ #
568
+ warning = warning or 8
569
+ critical = critical or 16
570
+ try:
571
+ data = get_server_status(con)
572
+ if not data['mem']['supported']:
573
+ print "OK - Platform not supported for memory info"
574
+ return 0
575
+ #
576
+ # convert to gigs
577
+ #
578
+ message = "Memory Usage:"
579
+ try:
580
+ mem_mapped = float(data['mem']['mapped']) / 1024.0
581
+ message += " %.2fGB mapped," % mem_mapped
582
+ except:
583
+ mem_mapped = -1
584
+ message += " mapped unsupported,"
585
+ try:
586
+ mem_mapped_journal = float(data['mem']['mappedWithJournal']) / 1024.0
587
+ message += " %.2fGB mappedWithJournal" % mem_mapped_journal
588
+ except:
589
+ mem_mapped_journal = 0
590
+ message += performance_data(perf_data, [("%.2f" % mem_mapped, "memory_mapped"), ("%.2f" % mem_mapped_journal, "mappedWithJournal")])
591
+
592
+ if not mem_mapped == -1:
593
+ return check_levels(mem_mapped, warning, critical, message)
594
+ else:
595
+ print "OK - Server does not provide mem.mapped info"
596
+ return 0
597
+
598
+ except Exception, e:
599
+ return exit_with_general_critical(e)
600
+
601
+
602
+ def check_lock(con, warning, critical, perf_data):
603
+ warning = warning or 10
604
+ critical = critical or 30
605
+ try:
606
+ data = get_server_status(con)
607
+ #
608
+ # calculate percentage
609
+ #
610
+ lock_percentage = float(data['globalLock']['lockTime']) / float(data['globalLock']['totalTime']) * 100
611
+ message = "Lock Percentage: %.2f%%" % lock_percentage
612
+ message += performance_data(perf_data, [("%.2f" % lock_percentage, "lock_percentage", warning, critical)])
613
+ return check_levels(lock_percentage, warning, critical, message)
614
+
615
+ except Exception, e:
616
+ return exit_with_general_critical(e)
617
+
618
+
619
+ def check_flushing(con, warning, critical, avg, perf_data):
620
+ #
621
+ # These thresholds mean it's taking 5 seconds to perform a background flush to issue a warning
622
+ # and 10 seconds to issue a critical.
623
+ #
624
+ warning = warning or 5000
625
+ critical = critical or 15000
626
+ try:
627
+ data = get_server_status(con)
628
+ if avg:
629
+ flush_time = float(data['backgroundFlushing']['average_ms'])
630
+ stat_type = "Average"
631
+ else:
632
+ flush_time = float(data['backgroundFlushing']['last_ms'])
633
+ stat_type = "Last"
634
+
635
+ message = "%s Flush Time: %.2fms" % (stat_type, flush_time)
636
+ message += performance_data(perf_data, [("%.2fms" % flush_time, "%s_flush_time" % stat_type.lower(), warning, critical)])
637
+
638
+ return check_levels(flush_time, warning, critical, message)
639
+
640
+ except Exception, e:
641
+ return exit_with_general_critical(e)
642
+
643
+
644
+ def index_miss_ratio(con, warning, critical, perf_data):
645
+ warning = warning or 10
646
+ critical = critical or 30
647
+ try:
648
+ data = get_server_status(con)
649
+
650
+ try:
651
+ serverVersion = tuple(con.server_info()['version'].split('.'))
652
+ if serverVersion >= tuple("2.4.0".split(".")):
653
+ miss_ratio = float(data['indexCounters']['missRatio'])
654
+ else:
655
+ miss_ratio = float(data['indexCounters']['btree']['missRatio'])
656
+ except KeyError:
657
+ not_supported_msg = "not supported on this platform"
658
+ if data['indexCounters'].has_key('note'):
659
+ print "OK - MongoDB says: " + not_supported_msg
660
+ return 0
661
+ else:
662
+ print "WARNING - Can't get counter from MongoDB"
663
+ return 1
664
+
665
+ message = "Miss Ratio: %.2f" % miss_ratio
666
+ message += performance_data(perf_data, [("%.2f" % miss_ratio, "index_miss_ratio", warning, critical)])
667
+
668
+ return check_levels(miss_ratio, warning, critical, message)
669
+
670
+ except Exception, e:
671
+ return exit_with_general_critical(e)
672
+
673
+ def check_replset_quorum(con, perf_data):
674
+ db = con['admin']
675
+ warning = 1
676
+ critical = 2
677
+ primary = 0
678
+
679
+ try:
680
+ rs_members = db.command("replSetGetStatus")['members']
681
+
682
+ for member in rs_members:
683
+ if member['state'] == 1:
684
+ primary += 1
685
+
686
+ if primary == 1:
687
+ state = 0
688
+ message = "Cluster is quorate"
689
+ else:
690
+ state = 2
691
+ message = "Cluster is not quorate and cannot operate"
692
+
693
+ return check_levels(state, warning, critical, message)
694
+ except Exception, e:
695
+ return exit_with_general_critical(e)
696
+
697
+
698
+
699
+ def check_replset_state(con, perf_data, warning="", critical=""):
700
+ try:
701
+ warning = [int(x) for x in warning.split(",")]
702
+ except:
703
+ warning = [0, 3, 5]
704
+ try:
705
+ critical = [int(x) for x in critical.split(",")]
706
+ except:
707
+ critical = [8, 4, -1]
708
+
709
+ ok = range(-1, 8) # should include the range of all posiible values
710
+ try:
711
+ try:
712
+ try:
713
+ set_read_preference(con.admin)
714
+ data = con.admin.command(pymongo.son_manipulator.SON([('replSetGetStatus', 1)]))
715
+ except:
716
+ data = con.admin.command(son.SON([('replSetGetStatus', 1)]))
717
+ state = int(data['myState'])
718
+ except pymongo.errors.OperationFailure, e:
719
+ if e.code == None and str(e).find('failed: not running with --replSet"'):
720
+ state = -1
721
+
722
+ if state == 8:
723
+ message = "State: %i (Down)" % state
724
+ elif state == 4:
725
+ message = "State: %i (Fatal error)" % state
726
+ elif state == 0:
727
+ message = "State: %i (Starting up, phase1)" % state
728
+ elif state == 3:
729
+ message = "State: %i (Recovering)" % state
730
+ elif state == 5:
731
+ message = "State: %i (Starting up, phase2)" % state
732
+ elif state == 1:
733
+ message = "State: %i (Primary)" % state
734
+ elif state == 2:
735
+ message = "State: %i (Secondary)" % state
736
+ elif state == 7:
737
+ message = "State: %i (Arbiter)" % state
738
+ elif state == -1:
739
+ message = "Not running with replSet"
740
+ else:
741
+ message = "State: %i (Unknown state)" % state
742
+ message += performance_data(perf_data, [(state, "state")])
743
+ return check_levels(state, warning, critical, message, ok)
744
+ except Exception, e:
745
+ return exit_with_general_critical(e)
746
+
747
+
748
+ def check_databases(con, warning, critical, perf_data=None):
749
+ try:
750
+ try:
751
+ set_read_preference(con.admin)
752
+ data = con.admin.command(pymongo.son_manipulator.SON([('listDatabases', 1)]))
753
+ except:
754
+ data = con.admin.command(son.SON([('listDatabases', 1)]))
755
+
756
+ count = len(data['databases'])
757
+ message = "Number of DBs: %.0f" % count
758
+ message += performance_data(perf_data, [(count, "databases", warning, critical, message)])
759
+ return check_levels(count, warning, critical, message)
760
+ except Exception, e:
761
+ return exit_with_general_critical(e)
762
+
763
+
764
+ def check_collections(con, warning, critical, perf_data=None):
765
+ try:
766
+ try:
767
+ set_read_preference(con.admin)
768
+ data = con.admin.command(pymongo.son_manipulator.SON([('listDatabases', 1)]))
769
+ except:
770
+ data = con.admin.command(son.SON([('listDatabases', 1)]))
771
+
772
+ count = 0
773
+ for db in data['databases']:
774
+ dbase = con[db['name']]
775
+ set_read_preference(dbase)
776
+ count += len(dbase.collection_names())
777
+
778
+ message = "Number of collections: %.0f" % count
779
+ message += performance_data(perf_data, [(count, "collections", warning, critical, message)])
780
+ return check_levels(count, warning, critical, message)
781
+
782
+ except Exception, e:
783
+ return exit_with_general_critical(e)
784
+
785
+
786
+ def check_all_databases_size(con, warning, critical, perf_data):
787
+ warning = warning or 100
788
+ critical = critical or 1000
789
+ try:
790
+ set_read_preference(con.admin)
791
+ all_dbs_data = con.admin.command(pymongo.son_manipulator.SON([('listDatabases', 1)]))
792
+ except:
793
+ all_dbs_data = con.admin.command(son.SON([('listDatabases', 1)]))
794
+
795
+ total_storage_size = 0
796
+ message = ""
797
+ perf_data_param = [()]
798
+ for db in all_dbs_data['databases']:
799
+ database = db['name']
800
+ data = con[database].command('dbstats')
801
+ storage_size = round(data['storageSize'] / 1024 / 1024, 1)
802
+ message += "; Database %s size: %.0f MB" % (database, storage_size)
803
+ perf_data_param.append((storage_size, database + "_database_size"))
804
+ total_storage_size += storage_size
805
+
806
+ perf_data_param[0] = (total_storage_size, "total_size", warning, critical)
807
+ message += performance_data(perf_data, perf_data_param)
808
+ message = "Total size: %.0f MB" % total_storage_size + message
809
+ return check_levels(total_storage_size, warning, critical, message)
810
+
811
+
812
+ def check_database_size(con, database, warning, critical, perf_data):
813
+ warning = warning or 100
814
+ critical = critical or 1000
815
+ perfdata = ""
816
+ try:
817
+ set_read_preference(con.admin)
818
+ data = con[database].command('dbstats')
819
+ storage_size = data['storageSize'] / 1024 / 1024
820
+ if perf_data:
821
+ perfdata += " | database_size=%i;%i;%i" % (storage_size, warning, critical)
822
+ #perfdata += " database=%s" %(database)
823
+
824
+ if storage_size >= critical:
825
+ print "CRITICAL - Database size: %.0f MB, Database: %s%s" % (storage_size, database, perfdata)
826
+ return 2
827
+ elif storage_size >= warning:
828
+ print "WARNING - Database size: %.0f MB, Database: %s%s" % (storage_size, database, perfdata)
829
+ return 1
830
+ else:
831
+ print "OK - Database size: %.0f MB, Database: %s%s" % (storage_size, database, perfdata)
832
+ return 0
833
+ except Exception, e:
834
+ return exit_with_general_critical(e)
835
+
836
+
837
+ def check_database_indexes(con, database, warning, critical, perf_data):
838
+ #
839
+ # These thresholds are basically meaningless, and must be customized to your application
840
+ #
841
+ warning = warning or 100
842
+ critical = critical or 1000
843
+ perfdata = ""
844
+ try:
845
+ set_read_preference(con.admin)
846
+ data = con[database].command('dbstats')
847
+ index_size = data['indexSize'] / 1024 / 1024
848
+ if perf_data:
849
+ perfdata += " | database_indexes=%i;%i;%i" % (index_size, warning, critical)
850
+
851
+ if index_size >= critical:
852
+ print "CRITICAL - %s indexSize: %.0f MB %s" % (database, index_size, perfdata)
853
+ return 2
854
+ elif index_size >= warning:
855
+ print "WARNING - %s indexSize: %.0f MB %s" % (database, index_size, perfdata)
856
+ return 1
857
+ else:
858
+ print "OK - %s indexSize: %.0f MB %s" % (database, index_size, perfdata)
859
+ return 0
860
+ except Exception, e:
861
+ return exit_with_general_critical(e)
862
+
863
+
864
+ def check_collection_indexes(con, database, collection, warning, critical, perf_data):
865
+ #
866
+ # These thresholds are basically meaningless, and must be customized to your application
867
+ #
868
+ warning = warning or 100
869
+ critical = critical or 1000
870
+ perfdata = ""
871
+ try:
872
+ set_read_preference(con.admin)
873
+ data = con[database].command('collstats', collection)
874
+ total_index_size = data['totalIndexSize'] / 1024 / 1024
875
+ if perf_data:
876
+ perfdata += " | collection_indexes=%i;%i;%i" % (total_index_size, warning, critical)
877
+
878
+ if total_index_size >= critical:
879
+ print "CRITICAL - %s.%s totalIndexSize: %.0f MB %s" % (database, collection, total_index_size, perfdata)
880
+ return 2
881
+ elif total_index_size >= warning:
882
+ print "WARNING - %s.%s totalIndexSize: %.0f MB %s" % (database, collection, total_index_size, perfdata)
883
+ return 1
884
+ else:
885
+ print "OK - %s.%s totalIndexSize: %.0f MB %s" % (database, collection, total_index_size, perfdata)
886
+ return 0
887
+ except Exception, e:
888
+ return exit_with_general_critical(e)
889
+
890
+
891
+ def check_queues(con, warning, critical, perf_data):
892
+ warning = warning or 10
893
+ critical = critical or 30
894
+ try:
895
+ data = get_server_status(con)
896
+
897
+ total_queues = float(data['globalLock']['currentQueue']['total'])
898
+ readers_queues = float(data['globalLock']['currentQueue']['readers'])
899
+ writers_queues = float(data['globalLock']['currentQueue']['writers'])
900
+ message = "Current queue is : total = %d, readers = %d, writers = %d" % (total_queues, readers_queues, writers_queues)
901
+ message += performance_data(perf_data, [(total_queues, "total_queues", warning, critical), (readers_queues, "readers_queues"), (writers_queues, "writers_queues")])
902
+ return check_levels(total_queues, warning, critical, message)
903
+
904
+ except Exception, e:
905
+ return exit_with_general_critical(e)
906
+
907
+ def check_collection_size(con, database, collection, warning, critical, perf_data):
908
+ warning = warning or 100
909
+ critical = critical or 1000
910
+ perfdata = ""
911
+ try:
912
+ set_read_preference(con.admin)
913
+ data = con[database].command('collstats', collection)
914
+ size = data['size'] / 1024 / 1024
915
+ if perf_data:
916
+ perfdata += " | collection_size=%i;%i;%i" % (size, warning, critical)
917
+
918
+ if size >= critical:
919
+ print "CRITICAL - %s.%s size: %.0f MB %s" % (database, collection, size, perfdata)
920
+ return 2
921
+ elif size >= warning:
922
+ print "WARNING - %s.%s size: %.0f MB %s" % (database, collection, size, perfdata)
923
+ return 1
924
+ else:
925
+ print "OK - %s.%s size: %.0f MB %s" % (database, collection, size, perfdata)
926
+ return 0
927
+ except Exception, e:
928
+ return exit_with_general_critical(e)
929
+
930
+ def check_queries_per_second(con, query_type, warning, critical, perf_data):
931
+ warning = warning or 250
932
+ critical = critical or 500
933
+
934
+ if query_type not in ['insert', 'query', 'update', 'delete', 'getmore', 'command']:
935
+ return exit_with_general_critical("The query type of '%s' is not valid" % query_type)
936
+
937
+ try:
938
+ db = con.local
939
+ data = get_server_status(con)
940
+
941
+ # grab the count
942
+ num = int(data['opcounters'][query_type])
943
+
944
+ # do the math
945
+ last_count = db.nagios_check.find_one({'check': 'query_counts'})
946
+ try:
947
+ ts = int(time.time())
948
+ diff_query = num - last_count['data'][query_type]['count']
949
+ diff_ts = ts - last_count['data'][query_type]['ts']
950
+
951
+ query_per_sec = float(diff_query) / float(diff_ts)
952
+
953
+ # update the count now
954
+ db.nagios_check.update({u'_id': last_count['_id']}, {'$set': {"data.%s" % query_type: {'count': num, 'ts': int(time.time())}}})
955
+
956
+ message = "Queries / Sec: %f" % query_per_sec
957
+ message += performance_data(perf_data, [(query_per_sec, "%s_per_sec" % query_type, warning, critical, message)])
958
+ except KeyError:
959
+ #
960
+ # since it is the first run insert it
961
+ query_per_sec = 0
962
+ message = "First run of check.. no data"
963
+ db.nagios_check.update({u'_id': last_count['_id']}, {'$set': {"data.%s" % query_type: {'count': num, 'ts': int(time.time())}}})
964
+ except TypeError:
965
+ #
966
+ # since it is the first run insert it
967
+ query_per_sec = 0
968
+ message = "First run of check.. no data"
969
+ db.nagios_check.insert({'check': 'query_counts', 'data': {query_type: {'count': num, 'ts': int(time.time())}}})
970
+
971
+ return check_levels(query_per_sec, warning, critical, message)
972
+
973
+ except Exception, e:
974
+ return exit_with_general_critical(e)
975
+
976
+
977
+ def check_oplog(con, warning, critical, perf_data):
978
+ """ Checking the oplog time - the time of the log currntly saved in the oplog collection
979
+ defaults:
980
+ critical 4 hours
981
+ warning 24 hours
982
+ those can be changed as usual with -C and -W parameters"""
983
+ warning = warning or 24
984
+ critical = critical or 4
985
+ try:
986
+ db = con.local
987
+ ol = db.system.namespaces.find_one({"name": "local.oplog.rs"})
988
+ if (db.system.namespaces.find_one({"name": "local.oplog.rs"}) != None):
989
+ oplog = "oplog.rs"
990
+ else:
991
+ ol = db.system.namespaces.find_one({"name": "local.oplog.$main"})
992
+ if (db.system.namespaces.find_one({"name": "local.oplog.$main"}) != None):
993
+ oplog = "oplog.$main"
994
+ else:
995
+ message = "neither master/slave nor replica set replication detected"
996
+ return check_levels(None, warning, critical, message)
997
+
998
+ try:
999
+ set_read_preference(con.admin)
1000
+ data = con.local.command(pymongo.son_manipulator.SON([('collstats', oplog)]))
1001
+ except:
1002
+ data = con.admin.command(son.SON([('collstats', oplog)]))
1003
+
1004
+ ol_size = data['size']
1005
+ ol_storage_size = data['storageSize']
1006
+ ol_used_storage = int(float(ol_size) / ol_storage_size * 100 + 1)
1007
+ ol = con.local[oplog]
1008
+ firstc = ol.find().sort("$natural", pymongo.ASCENDING).limit(1)[0]['ts']
1009
+ lastc = ol.find().sort("$natural", pymongo.DESCENDING).limit(1)[0]['ts']
1010
+ time_in_oplog = (lastc.as_datetime() - firstc.as_datetime())
1011
+ message = "Oplog saves " + str(time_in_oplog) + " %d%% used" % ol_used_storage
1012
+ try: # work starting from python2.7
1013
+ hours_in_oplog = time_in_oplog.total_seconds() / 60 / 60
1014
+ except:
1015
+ hours_in_oplog = float(time_in_oplog.seconds + time_in_oplog.days * 24 * 3600) / 60 / 60
1016
+ approx_level = hours_in_oplog * 100 / ol_used_storage
1017
+ message += performance_data(perf_data, [("%.2f" % hours_in_oplog, 'oplog_time', warning, critical), ("%.2f " % approx_level, 'oplog_time_100_percent_used')])
1018
+ return check_levels(-approx_level, -warning, -critical, message)
1019
+
1020
+ except Exception, e:
1021
+ return exit_with_general_critical(e)
1022
+
1023
+
1024
+ def check_journal_commits_in_wl(con, warning, critical, perf_data):
1025
+ """ Checking the number of commits which occurred in the db's write lock.
1026
+ Most commits are performed outside of this lock; committed while in the write lock is undesirable.
1027
+ Under very high write situations it is normal for this value to be nonzero. """
1028
+
1029
+ warning = warning or 10
1030
+ critical = critical or 40
1031
+ try:
1032
+ data = get_server_status(con)
1033
+ j_commits_in_wl = data['dur']['commitsInWriteLock']
1034
+ message = "Journal commits in DB write lock : %d" % j_commits_in_wl
1035
+ message += performance_data(perf_data, [(j_commits_in_wl, "j_commits_in_wl", warning, critical)])
1036
+ return check_levels(j_commits_in_wl, warning, critical, message)
1037
+
1038
+ except Exception, e:
1039
+ return exit_with_general_critical(e)
1040
+
1041
+
1042
+ def check_journaled(con, warning, critical, perf_data):
1043
+ """ Checking the average amount of data in megabytes written to the recovery log in the last four seconds"""
1044
+
1045
+ warning = warning or 20
1046
+ critical = critical or 40
1047
+ try:
1048
+ data = get_server_status(con)
1049
+ journaled = data['dur']['journaledMB']
1050
+ message = "Journaled : %.2f MB" % journaled
1051
+ message += performance_data(perf_data, [("%.2f" % journaled, "journaled", warning, critical)])
1052
+ return check_levels(journaled, warning, critical, message)
1053
+
1054
+ except Exception, e:
1055
+ return exit_with_general_critical(e)
1056
+
1057
+
1058
+ def check_write_to_datafiles(con, warning, critical, perf_data):
1059
+ """ Checking the average amount of data in megabytes written to the databases datafiles in the last four seconds.
1060
+ As these writes are already journaled, they can occur lazily, and thus the number indicated here may be lower
1061
+ than the amount physically written to disk."""
1062
+ warning = warning or 20
1063
+ critical = critical or 40
1064
+ try:
1065
+ data = get_server_status(con)
1066
+ writes = data['dur']['writeToDataFilesMB']
1067
+ message = "Write to data files : %.2f MB" % writes
1068
+ message += performance_data(perf_data, [("%.2f" % writes, "write_to_data_files", warning, critical)])
1069
+ return check_levels(writes, warning, critical, message)
1070
+
1071
+ except Exception, e:
1072
+ return exit_with_general_critical(e)
1073
+
1074
+
1075
+ def get_opcounters(data, opcounters_name, host):
1076
+ try:
1077
+ insert = data[opcounters_name]['insert']
1078
+ query = data[opcounters_name]['query']
1079
+ update = data[opcounters_name]['update']
1080
+ delete = data[opcounters_name]['delete']
1081
+ getmore = data[opcounters_name]['getmore']
1082
+ command = data[opcounters_name]['command']
1083
+ except KeyError, e:
1084
+ return 0, [0] * 100
1085
+ total_commands = insert + query + update + delete + getmore + command
1086
+ new_vals = [total_commands, insert, query, update, delete, getmore, command]
1087
+ return maintain_delta(new_vals, host, opcounters_name)
1088
+
1089
+
1090
+ def check_opcounters(con, host, warning, critical, perf_data):
1091
+ """ A function to get all opcounters delta per minute. In case of a replication - gets the opcounters+opcountersRepl"""
1092
+ warning = warning or 10000
1093
+ critical = critical or 15000
1094
+
1095
+ data = get_server_status(con)
1096
+ err1, delta_opcounters = get_opcounters(data, 'opcounters', host)
1097
+ err2, delta_opcounters_repl = get_opcounters(data, 'opcountersRepl', host)
1098
+ if err1 == 0 and err2 == 0:
1099
+ delta = [(x + y) for x, y in zip(delta_opcounters, delta_opcounters_repl)]
1100
+ delta[0] = delta_opcounters[0] # only the time delta shouldn't be summarized
1101
+ per_minute_delta = [int(x / delta[0] * 60) for x in delta[1:]]
1102
+ message = "Test succeeded , old values missing"
1103
+ message = "Opcounters: total=%d,insert=%d,query=%d,update=%d,delete=%d,getmore=%d,command=%d" % tuple(per_minute_delta)
1104
+ message += performance_data(perf_data, ([(per_minute_delta[0], "total", warning, critical), (per_minute_delta[1], "insert"),
1105
+ (per_minute_delta[2], "query"), (per_minute_delta[3], "update"), (per_minute_delta[5], "delete"),
1106
+ (per_minute_delta[5], "getmore"), (per_minute_delta[6], "command")]))
1107
+ return check_levels(per_minute_delta[0], warning, critical, message)
1108
+ else:
1109
+ return exit_with_general_critical("problem reading data from temp file")
1110
+
1111
+
1112
+ def check_current_lock(con, host, warning, critical, perf_data):
1113
+ """ A function to get current lock percentage and not a global one, as check_lock function does"""
1114
+ warning = warning or 10
1115
+ critical = critical or 30
1116
+ data = get_server_status(con)
1117
+
1118
+ lockTime = float(data['globalLock']['lockTime'])
1119
+ totalTime = float(data['globalLock']['totalTime'])
1120
+
1121
+ err, delta = maintain_delta([totalTime, lockTime], host, "locktime")
1122
+ if err == 0:
1123
+ lock_percentage = delta[2] / delta[1] * 100 # lockTime/totalTime*100
1124
+ message = "Current Lock Percentage: %.2f%%" % lock_percentage
1125
+ message += performance_data(perf_data, [("%.2f" % lock_percentage, "current_lock_percentage", warning, critical)])
1126
+ return check_levels(lock_percentage, warning, critical, message)
1127
+ else:
1128
+ return exit_with_general_warning("problem reading data from temp file")
1129
+
1130
+
1131
+ def check_page_faults(con, host, warning, critical, perf_data):
1132
+ """ A function to get page_faults per second from the system"""
1133
+ warning = warning or 10
1134
+ critical = critical or 30
1135
+ data = get_server_status(con)
1136
+
1137
+ try:
1138
+ page_faults = float(data['extra_info']['page_faults'])
1139
+ except:
1140
+ # page_faults unsupported on the underlaying system
1141
+ return exit_with_general_critical("page_faults unsupported on the underlaying system")
1142
+
1143
+ err, delta = maintain_delta([page_faults], host, "page_faults")
1144
+ if err == 0:
1145
+ page_faults_ps = delta[1] / delta[0]
1146
+ message = "Page faults : %.2f ps" % page_faults_ps
1147
+ message += performance_data(perf_data, [("%.2f" % page_faults_ps, "page_faults_ps", warning, critical)])
1148
+ return check_levels(page_faults_ps, warning, critical, message)
1149
+ else:
1150
+ return exit_with_general_warning("problem reading data from temp file")
1151
+
1152
+
1153
+ def check_asserts(con, host, warning, critical, perf_data):
1154
+ """ A function to get asserts from the system"""
1155
+ warning = warning or 1
1156
+ critical = critical or 10
1157
+ data = get_server_status(con)
1158
+
1159
+ asserts = data['asserts']
1160
+
1161
+ #{ "regular" : 0, "warning" : 6, "msg" : 0, "user" : 12, "rollovers" : 0 }
1162
+ regular = asserts['regular']
1163
+ warning_asserts = asserts['warning']
1164
+ msg = asserts['msg']
1165
+ user = asserts['user']
1166
+ rollovers = asserts['rollovers']
1167
+
1168
+ err, delta = maintain_delta([regular, warning_asserts, msg, user, rollovers], host, "asserts")
1169
+
1170
+ if err == 0:
1171
+ if delta[5] != 0:
1172
+ #the number of rollovers were increased
1173
+ warning = -1 # no matter the metrics this situation should raise a warning
1174
+ # if this is normal rollover - the warning will not appear again, but if there will be a lot of asserts
1175
+ # the warning will stay for a long period of time
1176
+ # although this is not a usual situation
1177
+
1178
+ regular_ps = delta[1] / delta[0]
1179
+ warning_ps = delta[2] / delta[0]
1180
+ msg_ps = delta[3] / delta[0]
1181
+ user_ps = delta[4] / delta[0]
1182
+ rollovers_ps = delta[5] / delta[0]
1183
+ total_ps = regular_ps + warning_ps + msg_ps + user_ps
1184
+ message = "Total asserts : %.2f ps" % total_ps
1185
+ message += performance_data(perf_data, [(total_ps, "asserts_ps", warning, critical), (regular_ps, "regular"),
1186
+ (warning_ps, "warning"), (msg_ps, "msg"), (user_ps, "user")])
1187
+ return check_levels(total_ps, warning, critical, message)
1188
+ else:
1189
+ return exit_with_general_warning("problem reading data from temp file")
1190
+
1191
+
1192
+ def get_stored_primary_server_name(db):
1193
+ """ get the stored primary server name from db. """
1194
+ if "last_primary_server" in db.collection_names():
1195
+ stored_primary_server = db.last_primary_server.find_one()["server"]
1196
+ else:
1197
+ stored_primary_server = None
1198
+
1199
+ return stored_primary_server
1200
+
1201
+
1202
+ def check_replica_primary(con, host, warning, critical, perf_data, replicaset):
1203
+ """ A function to check if the primary server of a replica set has changed """
1204
+ if warning is None and critical is None:
1205
+ warning = 1
1206
+ warning = warning or 2
1207
+ critical = critical or 2
1208
+
1209
+ primary_status = 0
1210
+ message = "Primary server has not changed"
1211
+ db = con["nagios"]
1212
+ data = get_server_status(con)
1213
+ if replicaset != data['repl'].get('setName'):
1214
+ message = "Replica set requested: %s differs from the one found: %s" % (replicaset, data['repl'].get('setName'))
1215
+ primary_status = 2
1216
+ return check_levels(primary_status, warning, critical, message)
1217
+ current_primary = data['repl'].get('primary')
1218
+ saved_primary = get_stored_primary_server_name(db)
1219
+ if current_primary is None:
1220
+ current_primary = "None"
1221
+ if saved_primary is None:
1222
+ saved_primary = "None"
1223
+ if current_primary != saved_primary:
1224
+ last_primary_server_record = {"server": current_primary}
1225
+ db.last_primary_server.update({"_id": "last_primary"}, {"$set": last_primary_server_record}, upsert=True, safe=True)
1226
+ message = "Primary server has changed from %s to %s" % (saved_primary, current_primary)
1227
+ primary_status = 1
1228
+ return check_levels(primary_status, warning, critical, message)
1229
+
1230
+
1231
+ def check_page_faults(con, sample_time, warning, critical, perf_data):
1232
+ warning = warning or 10
1233
+ critical = critical or 20
1234
+ try:
1235
+ try:
1236
+ set_read_preference(con.admin)
1237
+ data1 = con.admin.command(pymongo.son_manipulator.SON([('serverStatus', 1)]))
1238
+ time.sleep(sample_time)
1239
+ data2 = con.admin.command(pymongo.son_manipulator.SON([('serverStatus', 1)]))
1240
+ except:
1241
+ data1 = con.admin.command(son.SON([('serverStatus', 1)]))
1242
+ time.sleep(sample_time)
1243
+ data2 = con.admin.command(son.SON([('serverStatus', 1)]))
1244
+
1245
+ try:
1246
+ #on linux servers only
1247
+ page_faults = (int(data2['extra_info']['page_faults']) - int(data1['extra_info']['page_faults'])) / sample_time
1248
+ except KeyError:
1249
+ print "WARNING - Can't get extra_info.page_faults counter from MongoDB"
1250
+ sys.exit(1)
1251
+
1252
+ message = "Page Faults: %i" % (page_faults)
1253
+
1254
+ message += performance_data(perf_data, [(page_faults, "page_faults", warning, critical)])
1255
+ check_levels(page_faults, warning, critical, message)
1256
+
1257
+ except Exception, e:
1258
+ exit_with_general_critical(e)
1259
+
1260
+
1261
+ def chunks_balance(con, database, collection, warning, critical):
1262
+ warning = warning or 10
1263
+ critical = critical or 20
1264
+ nsfilter = database + "." + collection
1265
+ try:
1266
+ try:
1267
+ set_read_preference(con.admin)
1268
+ col = con.config.chunks
1269
+ nscount = col.find({"ns": nsfilter}).count()
1270
+ shards = col.distinct("shard")
1271
+
1272
+ except:
1273
+ print "WARNING - Can't get chunks infos from MongoDB"
1274
+ sys.exit(1)
1275
+
1276
+ if nscount == 0:
1277
+ print "WARNING - Namespace %s is not sharded" % (nsfilter)
1278
+ sys.exit(1)
1279
+
1280
+ avgchunksnb = nscount / len(shards)
1281
+ warningnb = avgchunksnb * warning / 100
1282
+ criticalnb = avgchunksnb * critical / 100
1283
+
1284
+ for shard in shards:
1285
+ delta = abs(avgchunksnb - col.find({"ns": nsfilter, "shard": shard}).count())
1286
+ message = "Namespace: %s, Shard name: %s, Chunk delta: %i" % (nsfilter, shard, delta)
1287
+
1288
+ if delta >= criticalnb and delta > 0:
1289
+ print "CRITICAL - Chunks not well balanced " + message
1290
+ sys.exit(2)
1291
+ elif delta >= warningnb and delta > 0:
1292
+ print "WARNING - Chunks not well balanced " + message
1293
+ sys.exit(1)
1294
+
1295
+ print "OK - Chunks well balanced across shards"
1296
+ sys.exit(0)
1297
+
1298
+ except Exception, e:
1299
+ exit_with_general_critical(e)
1300
+
1301
+ print "OK - Chunks well balanced across shards"
1302
+ sys.exit(0)
1303
+
1304
+
1305
+ def check_connect_primary(con, warning, critical, perf_data):
1306
+ warning = warning or 3
1307
+ critical = critical or 6
1308
+
1309
+ try:
1310
+ try:
1311
+ set_read_preference(con.admin)
1312
+ data = con.admin.command(pymongo.son_manipulator.SON([('isMaster', 1)]))
1313
+ except:
1314
+ data = con.admin.command(son.SON([('isMaster', 1)]))
1315
+
1316
+ if data['ismaster'] == True:
1317
+ print "OK - This server is primary"
1318
+ return 0
1319
+
1320
+ phost = data['primary'].split(':')[0]
1321
+ pport = int(data['primary'].split(':')[1])
1322
+ start = time.time()
1323
+
1324
+ err, con = mongo_connect(phost, pport)
1325
+ if err != 0:
1326
+ return err
1327
+
1328
+ pconn_time = time.time() - start
1329
+ pconn_time = round(pconn_time, 0)
1330
+ message = "Connection to primary server " + data['primary'] + " took %i seconds" % pconn_time
1331
+ message += performance_data(perf_data, [(pconn_time, "connection_time", warning, critical)])
1332
+
1333
+ return check_levels(pconn_time, warning, critical, message)
1334
+
1335
+ except Exception, e:
1336
+ return exit_with_general_critical(e)
1337
+
1338
+
1339
+ def check_collection_state(con, database, collection):
1340
+ try:
1341
+ con[database][collection].find_one()
1342
+ print "OK - Collection %s.%s is reachable " % (database, collection)
1343
+ return 0
1344
+
1345
+ except Exception, e:
1346
+ return exit_with_general_critical(e)
1347
+
1348
+
1349
+ def check_row_count(con, database, collection, warning, critical, perf_data):
1350
+ try:
1351
+ count = con[database][collection].count()
1352
+ message = "Row count: %i" % (count)
1353
+ message += performance_data(perf_data, [(count, "row_count", warning, critical)])
1354
+
1355
+ return check_levels(count, warning, critical, message)
1356
+
1357
+ except Exception, e:
1358
+ return exit_with_general_critical(e)
1359
+
1360
+
1361
+ def build_file_name(host, action):
1362
+ #done this way so it will work when run independently and from shell
1363
+ module_name = re.match('(.*//*)*(.*)\..*', __file__).group(2)
1364
+ return "/tmp/" + module_name + "_data/" + host + "-" + action + ".data"
1365
+
1366
+
1367
+ def ensure_dir(f):
1368
+ d = os.path.dirname(f)
1369
+ if not os.path.exists(d):
1370
+ os.makedirs(d)
1371
+
1372
+
1373
+ def write_values(file_name, string):
1374
+ f = None
1375
+ try:
1376
+ f = open(file_name, 'w')
1377
+ except IOError, e:
1378
+ #try creating
1379
+ if (e.errno == 2):
1380
+ ensure_dir(file_name)
1381
+ f = open(file_name, 'w')
1382
+ else:
1383
+ raise IOError(e)
1384
+ f.write(string)
1385
+ f.close()
1386
+ return 0
1387
+
1388
+
1389
+ def read_values(file_name):
1390
+ data = None
1391
+ try:
1392
+ f = open(file_name, 'r')
1393
+ data = f.read()
1394
+ f.close()
1395
+ return 0, data
1396
+ except IOError, e:
1397
+ if (e.errno == 2):
1398
+ #no previous data
1399
+ return 1, ''
1400
+ except Exception, e:
1401
+ return 2, None
1402
+
1403
+
1404
+ def calc_delta(old, new):
1405
+ delta = []
1406
+ if (len(old) != len(new)):
1407
+ raise Exception("unequal number of parameters")
1408
+ for i in range(0, len(old)):
1409
+ val = float(new[i]) - float(old[i])
1410
+ if val < 0:
1411
+ val = new[i]
1412
+ delta.append(val)
1413
+ return 0, delta
1414
+
1415
+
1416
+ def maintain_delta(new_vals, host, action):
1417
+ file_name = build_file_name(host, action)
1418
+ err, data = read_values(file_name)
1419
+ old_vals = data.split(';')
1420
+ new_vals = [str(int(time.time()))] + new_vals
1421
+ delta = None
1422
+ try:
1423
+ err, delta = calc_delta(old_vals, new_vals)
1424
+ except:
1425
+ err = 2
1426
+ write_res = write_values(file_name, ";" . join(str(x) for x in new_vals))
1427
+ return err + write_res, delta
1428
+
1429
+
1430
+ def replication_get_time_diff(con):
1431
+ col = 'oplog.rs'
1432
+ local = con.local
1433
+ ol = local.system.namespaces.find_one({"name": "local.oplog.$main"})
1434
+ if ol:
1435
+ col = 'oplog.$main'
1436
+ firstc = local[col].find().sort("$natural", 1).limit(1)
1437
+ lastc = local[col].find().sort("$natural", -1).limit(1)
1438
+ first = firstc.next()
1439
+ last = lastc.next()
1440
+ tfirst = first["ts"]
1441
+ tlast = last["ts"]
1442
+ delta = tlast.time - tfirst.time
1443
+ return delta
1444
+
1445
+ #
1446
+ # main app
1447
+ #
1448
+ if __name__ == "__main__":
1449
+ sys.exit(main(sys.argv[1:]))