@@ -49,6 +49,12 @@ affecting your performance, refer to the :ref:`server-status-locks`
49
49
section and the :ref:`globalLock` section of the
50
50
:dbcommand:`serverStatus` output.
51
51
52
+ .. note::
53
+ Some ``serverStatus`` response fields are not returned on
54
+ {+atlas+} M0/M2/M5 clusters. For more information, see
55
+ :ref:`free-shard-commands-with-limits` in the {+atlas+}
56
+ documentation.
57
+
52
58
Dividing :serverstatus:`locks.<type>.timeAcquiringMicros` by
53
59
:serverstatus:`locks.<type>.acquireWaitCount`
54
60
can give an approximate average wait time for a particular lock mode.
@@ -102,158 +108,27 @@ reuse connections more efficiently. An extremely high number of
102
108
connections, particularly without corresponding workload, is often
103
109
indicative of a driver or other configuration error.
104
110
111
+ Self-Managed Connection Limits
112
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
113
+
105
114
Unless constrained by system-wide limits, the maximum number of
106
115
incoming connections supported by MongoDB is configured with the
107
116
:setting:`~net.maxIncomingConnections` setting. On Unix-based systems,
108
117
system-wide limits can be modified using the ``ulimit`` command, or by
109
118
editing your system's ``/etc/sysctl`` file. See :doc:`/reference/ulimit`
110
119
for more information.
111
120
112
- .. _database-profiling:
113
-
114
- Database Profiling
115
- ------------------
116
-
117
- The :doc:`/tutorial/manage-the-database-profiler` collects detailed
118
- information about operations run against a mongod instance. The
119
- profiler's output can help to identify inefficient queries and
120
- operations.
121
-
122
- You can enable and configure profiling for individual databases or for
123
- all databases on a :binary:`~bin.mongod` instance.
124
- Profiler settings affect only a single :binary:`~bin.mongod` instance and
125
- will not propagate across a :term:`replica set` or :term:`sharded
126
- cluster`.
127
-
128
- See :doc:`/tutorial/manage-the-database-profiler` for information on
129
- enabling and configuring the profiler.
130
-
131
- The following profiling levels are available:
132
-
133
- .. include:: /includes/database-profiler-levels.rst
134
-
135
- .. include:: /includes/warning-profiler-performance.rst
136
-
137
- .. note::
138
-
139
- .. include:: /includes/fact-log-slow-queries.rst
140
-
141
- .. include:: /includes/extracts/4.2-changes-log-query-shapes-plan-cache-key.rst
142
-
143
- .. _ftdc-stub:
144
-
145
- Full Time Diagnostic Data Capture
146
- ---------------------------------
147
-
148
- To help MongoDB engineers analyze server behavior, :binary:`~bin.mongod` and
149
- :binary:`~bin.mongos` processes include a Full Time Diagnostic Data
150
- Capture (FTDC) mechanism. FTDC is enabled by default. Due to its
151
- importance in debugging deployments, FTDC thread failures are fatal and
152
- stop the parent ``mongod`` or ``mongos`` process.
153
-
154
- .. important:: FTDC Privacy
155
-
156
- FTDC data files are compressed and not human-readable. MongoDB Inc.
157
- engineers cannot access FTDC data without explicit permission
158
- and assistance from system owners or operators.
159
-
160
- FTDC data **never** contains any of the following information:
161
-
162
- - Samples of queries, query predicates, or query results
163
- - Data sampled from any end-user collection or index
164
- - System or MongoDB user credentials or security certificates
165
-
166
- FTDC data contains certain host machine information such as
167
- hostnames, operating system information, and the options or settings
168
- used to start the :binary:`mongod <bin.mongod>` or
169
- :binary:`mongos <bin.mongos>`. This information may be
170
- considered protected or confidential by some organizations or
171
- regulatory bodies, but is not typically considered to be Personally
172
- Identifiable Information (PII). For clusters where these fields were
173
- configured with protected, confidential, or PII data, please notify
174
- MongoDB Inc. engineers before sending the FTDC data so appropriate
175
- measures can be taken.
176
-
177
- FTDC periodically collects statistics produced by the following
178
- commands:
179
-
180
- - :dbcommand:`serverStatus`
181
- - :dbcommand:`replSetGetStatus` (:binary:`mongod <bin.mongod>` only)
182
- - :dbcommand:`collStats` for the :data:`local.oplog.rs` collection (:binary:`mongod <bin.mongod>` only)
183
- - :dbcommand:`connPoolStats` (:binary:`mongos <bin.mongos>` only)
184
-
185
- Depending on the host operating system, the diagnostic data may include
186
- one or more of the following utilization statistics:
187
-
188
- - CPU utilization
189
- - Memory utilization
190
- - Disk utilization related to performance. FTDC does not include
191
- data related to storage capacity.
192
- - Network performance statistics. FTDC only captures metadata and
193
- does not capture or inspect any network packets.
194
-
195
- .. note::
196
-
197
- If the :binary:`~bin.mongod` process runs in a :term:`container`, FTDC
198
- reports utilization statistics from the perspective of the container
199
- instead of the host operating system. For example, if a the
200
- :binary:`~bin.mongod` runs in a container that is configured with RAM
201
- restrictions, FTDC calculates memory utilization against the container's
202
- RAM limit, as opposed to the host operating system's RAM limit.
203
-
204
- FTDC collects statistics produced by the following commands on
205
- file rotation or startup:
206
-
207
- - :dbcommand:`getCmdLineOpts`
208
- - :dbcommand:`buildInfo`
209
- - :dbcommand:`hostInfo`
210
-
211
- :binary:`mongod <bin.mongod>` processes store FTDC data files in a
212
- ``diagnostic.data`` directory under the instances
213
- :setting:`storage.dbPath`. All diagnostic data files are stored
214
- under this directory. For example, given a :setting:`~storage.dbPath`
215
- of ``/data/db``, the diagnostic data directory would be
216
- ``/data/db/diagnostic.data``.
217
-
218
- :binary:`mongos <bin.mongos>` processes store FTDC data files in a
219
- diagnostic directory relative to the :setting:`systemLog.path` log
220
- path setting. MongoDB truncates the logpath's file extension and
221
- concatenates ``diagnostic.data`` to the remaining name. For example,
222
- given a :setting:`~systemLog.path` setting of
223
- ``/var/log/mongodb/mongos.log``, the diagnostic data directory would be
224
- ``/var/log/mongodb/mongos.diagnostic.data``.
225
-
226
- FTDC runs with the following defaults:
227
-
228
- - Data capture every 1 second
229
- - 200MB maximum ``diagnostic.data`` folder size.
230
-
231
- These defaults are designed to provide useful data to MongoDB Inc.
232
- engineers with minimal impact on performance or storage size. These
233
- values only require modifications if requested by MongoDB Inc. engineers
234
- for specific diagnostic purposes.
235
-
236
- You can view the FTDC source code on the
237
- `MongoDB Github Repository <https://github.com/mongodb/mongo/tree/master/src/mongo/db/ftdc>`_.
238
- The ``ftdc_system_stats_*.ccp`` files specifically define any
239
- system-specific diagnostic data captured.
240
-
241
- To disable FTDC, start up the :binary:`mongod <bin.mongod>` or
242
- :binary:`mongos <bin.mongos>` with the
243
- ``diagnosticDataCollectionEnabled: false`` option specified to the
244
- :setting:`setParameter` setting in your configuration file:
245
121
246
- .. code-block:: yaml
122
+ {+atlas+} Connection Limits
123
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
247
124
248
- setParameter:
249
- diagnosticDataCollectionEnabled: false
125
+ {+atlas+} sets the limit for concurrent incoming connections based on
126
+ the cluster tier and class. To learn more, see :ref:`<connection-limits>`
127
+ in the Atlas documentation.
250
128
251
- Disabling FTDC may increase the time or resources required when
252
- analyzing or debugging issues with support from MongoDB Inc. engineers.
253
129
254
130
.. toctree::
255
131
:titlesonly:
256
132
:hidden:
257
133
258
134
Connection Pool </administration/connection-pool-overview>
259
-
0 commit comments