processes
|
array
|
Required
|
Contains objects that define the
mongos
and
mongod
instances
that Cloud Manager monitors. Each object defines a different instance.
|
processes[n].args2_6
|
object
|
Required
|
MongoDB configuration object for MongoDB versions 2.6 and later.
See also
Supported configuration options
.
|
processes[n].alias
|
string
|
Optional
|
Hostname alias (often a
DNS
CNAME) for the host on which the
process runs. If an alias is specified, the MongoDB Agent prefers
this alias over the hostname specified in
processes.hostname
when connecting to the host. You can also specify this alias in
replicaSets.host
and
sharding.configServer
.
|
processes[n].authSchemaVersion
|
integer
|
Required
|
Schema version of the user credentials for MongoDB database
users. This should match all other elements of the
processes
array that belong to the same cluster.
-
Cloud Manager accepts
3
and
5
for this parameter.
-
MongoDB 3.x and 4.x clusters default to
5
.
-
MongoDB 2.6 clusters default to
3
.
See also
Upgrade to SCRAM-SHA-1
in the MongoDB 3.0 release notes.
|
processes[n].backupRestoreUrl
|
string
|
Optional
|
Delivery
URL
for the restore. Cloud Manager sets this when creating a
restore.
See also
Automate Backup Restoration through the API
.
|
processes[n].cluster
|
string
|
Conditional
|
Name of the sharded cluster. Set this value to the same value in
the
sharding.name
parameter in the
sharding
array for
the
mongos
.
-
Required
for a
mongos
.
-
Not needed
for a
mongod
.
|
defaultRWConcern.defaultReadConcern.level
|
string
|
Optional
|
Consistency and isolation properties set for the data read from
replica sets and replica set shards. MongoDB Atlas accepts the following values:
-
“available”
-
“local”
-
“majority”
|
defaultRWConcern.defaultWriteConcern.j
|
boolean
|
Optional
|
Flag that indicates whether the write acknowledgement must be
written to the
on-disk journal.
|
defaultRWConcern.defaultWriteConcern.w
|
string
|
Optional
|
Desired number of mongod instances that must acknowledge a write
operation in a replica sets and replica set shards. MongoDB Atlas accepts the
following values:
-
Any number 0 or greater
-
“majority”
|
defaultRWConcern.defaultWriteConcern.wtimeout
|
number
|
Optional
|
Desired time limit for the write concern
expressed in milliseconds. Set this value when you set
defaultRWConcern.defaultWriteConcern.w
to a value greater
than
1
.
|
processes[n].disabled
|
Boolean
|
Optional
|
Flag that indicates if this process should be shut down. Set to
true
to shut down the process.
|
processes[n].featureCompatibilityVersion
|
string
|
Required
|
Version of MongoDB with which this process has feature
compatibility. Changing this value can enable or disable certain
features that persist data incompatible with MongoDB versions
earlier or later than the
featureCompatibilityVersion
you
choose.
-
Cloud Manager accepts
3.2
,
3.6
,
4.2
and
4.4
as parameter values. If you have an
existing deployment, Cloud Manager only accepts a
featureCompatibilityVersion
equal to or one release older
than the MongoDB version you deployed. To learn which of
these parameter values is supported for each MongoDB version,
and which features each of these values enable or disable,
see setFeatureCompatibilityVersion in the
MongoDB Manual.
-
Cloud Manager sets this parameter to match the MongoDB version for new
deployments.
-
Cloud Manager doesn’t automatically increment this parameter when you
upgrade a host from one MongoDB version to the next.
See also
setFeatureCompatibilityVersion
|
processes[n].hostname
|
string
|
Required
|
Name of the host that serves this process. This defaults to
localhost
.
|
processes[n].lastCompact
|
string
|
Optional
|
Timestamp in ISO 8601 date and time format in
UTC
when Cloud Manager last reclaimed free space on a
cluster’s disks. During certain operations, MongoDB might move
or delete data but it doesn’t free the currently unused space. Cloud Manager
reclaims the disk space in a rolling fashion across members of
the replica set or shards.
To reclaim this space:
-
Immediately, set this value to the current time as an
ISO 8601 timestamp.
-
Later, set this value to a future ISO 8601 timestamp. Cloud Manager
reclaims the space after the current time passes the provided
timestamp.
To remove any ambiguity as to when you intend to reclaim the
space on the cluster’s disks, specify a time zone with your
ISO 8601 timestamp. For example, to set
processes.lastCompact
to 28 January 2021 at 2:43:52 PM US Central Standard Time, use
"processes.lastCompact"
:
"2021-01-28T14:43:52-06:00"
|
processes[n].lastRestart
|
string
|
Optional
|
Timestamp in ISO 8601 date and time format in
UTC
when Cloud Manager last restarted this process. If you
set this parameter to the current timestamp, Cloud Manager forces a
restart of this process after you upload this configuration.
If you set this parameter for multiple processes in the same
cluster, the Cloud Manager restarts the selected processes in a rolling
fashion across members of the replica set or shards.
|
processes[n].lastResync
|
string
|
Optional
|
Timestamp in ISO 8601 date and time format in
UTC
of the last
initial sync
process that Cloud Manager performed on the node.
To trigger the init sync process on the node immediately, set
this value to the current time as an ISO 8601 timestamp.
Warning
Use this parameter with caution. During
initial sync,
Automation removes the entire contents of the node’s
dbPath
directory.
If you set this parameter:
-
On the secondary node, the MongoDB Agent checks whether the
specified timestamp is later than the time of the last resync,
and if confirmed, starts init sync on this node.
Example
To set
processes.lastResync
on the secondary node to 28
May 2021 at 2:43:52 PM US CentralStandard Time, use:
"processes.lastResync"
:
"2021-05-28T14:43:52-06:00"
.
If the MongoDB Agent confirms that this timestamp is later
than the recorded time of the last resync, it starts init
sync on the node.
-
On the primary node, the MongoDB Agent waits until you ask the
primary node to become the secondary with the
rs.stepDown()
method, and then starts init sync on
this node.
-
On all of the nodes in the same cluster, including the
primary, the MongoDB Agent checks whether the specified
timestamp is later than the time of the last resync, and if
confirmed, starts init sync on the secondary nodes in a
rolling fashion. The MongoDB Agent waits until you ask the
primary node to become the secondary with the
rs.stepDown()
method, and then starts init sync on
this node.
|
processes[n].lastKmipMasterKeyRotation
|
string
|
Optional
|
Timestamp in ISO 8601 date and time format in
UTC
when Cloud Manager last rotated the master
KMIP
key. If
you set this parameter to the current timestamp, Cloud Manager rotate the key after you upload this configuration.
|
processes[n].logRotate
|
object
|
Optional
|
MongoDB configuration object for rotating the MongoDB logs of a
process.
|
processes[n].logRotate.
numTotal
|
integer
|
Optional
|
Total number of log files that Cloud Manager retains. If you don’t set
this value, the total number of log files defaults to
0
.
Cloud Manager bases rotation on your other
processes.logRotate
settings.
|
processes[n].logRotate.
numUncompressed
|
integer
|
Optional
|
Maximum number of total log files to leave uncompressed,
including the current log file. The default is
5
.
|
processes[n].logRotate.
percentOfDiskspace
|
number
|
Optional
|
Maximum percentage of total disk space that Cloud Manager can use to
store the log files expressed as decimal. If this limit is
exceeded, Cloud Manager deletes compressed log files until it meets this
limit. Cloud Manager deletes the oldest log files first.
The default is
0.02
.
|
processes[n].logRotate.
sizeThresholdMB
|
number
|
Required
|
Maximum size in MB for an individual log file before Cloud Manager
rotates it. Cloud Manager rotates the log file immediately if it meets
the value given in either this
sizeThresholdMB
or the
processes.logRotate.timeThresholdHrs
limit.
|
processes[n].logRotate.
timeThresholdHrs
|
integer
|
Required
|
Maximum duration in hours for an individual log file before the
next rotation. The time is since the last rotation.
Cloud Manager rotates the log file once the file meets either this
timeThresholdHrs
or the
processes.logRotate.sizeThresholdMB
limit.
|
processes[n].manualMode
|
Boolean
|
Optional
|
Flag that indicates if MongoDB Agent automates this process.
-
This defaults to
false
.
-
Set to
true
to disable Automation on this process. The
MongoDB Agent takes no further actions on this process.
-
Set to
false
to enable Automation on this process. The
MongoDB Agent automates actions on this process.
|
processes[n].name
|
string
|
Required
|
Unique name to identify the instance.
|
processes[n].numCores
|
integer
|
Optional
|
Number of cores that Cloud Manager should bind to this process. The
MongoDB Agent distributes processes across the cores as evenly as
possible.
|
processes[n].processType
|
string
|
Required
|
Type of MongoDB process being run. Cloud Manager accepts
mongod
or
mongos
for this parameter.
|
processes[n].version
|
string
|
Required
|
Name of the
mongoDbVersions
specification used with this
instance.
|