mongo_replset

Search…

mongo_replset

The out_mongo_replset Output plugin writes records into MongoDB, the emerging document-oriented database system.
This plugin is for users using ReplicaSet. If you are not using ReplicaSet, please see the out_mongo article instead.
This plugin has breaking changes since 0.8.0 due to mongo-ruby driver's breaking changes. If you are using a prior 0.7.x series, please be careful to upgrade 1.0.0 or later versions.
Why Fluentd with MongoDB?
Fluentd enables your apps to insert records to MongoDB asynchronously with batch-insertion, unlike direct insertion of records from your apps. This has the following advantages:
1.
less impact on application performance
2.
higher MongoDB insertion throughput while maintaining JSON record structure
Install
out_mongo_replset is not included in td-agent, by default. Fluentd gem users will need to install the fluent-plugin-mongo gem using the following command:
1
$ fluent-gem install fluent-plugin-mongo
Copied!
Example Configuration
1
# Single MongoDB
2
<match mongo.**>
3
  @type mongo_replset
4
  database fluentd
5
  collection test
6
  nodes localhost:27017,localhost:27018,localhost:27019
7
​
8
  # The name of the replica set
9
  replica_set myapp
10
​
11
  <buffer>
12
    # flush
13
    flush_interval 10s
14
  </buffer>
15
</match>
Copied!
Please see the Store Apache Logs into MongoDB article for real-world use cases.
Please see the Configuration File article for the basic structure and syntax of the configuration file.
For <buffer>, refer to Buffer Section Configuration.
Parameters
@type
The value must be mongo.
nodes
type
default
version
string
required parameter
1.0.0
The comma-separated node strings (e.g. host1:27017,host2:27017,host3:27017).
database
type
default
version
string
required parameter
1.0.0
The database name.
collection
type
default
version
string
required parameter if not tag_mapped
1.0.0
The collection name.
capped
type
default
version
string
optional
1.0.0
This option enables the capped collection. This is always recommended because MongoDB is not suited to storing large amounts of historical data.
capped_size
type
default
version
size
optional
1.0.0
Sets the capped collection size.
user
type
default
version
string
optional
1.0.0
The username to use for authentication.
password
type
default
version
string
optional
1.0.0
The password to use for authentication.
tag_mapped
type
default
version
string
optional
1.0.0
This option allows out_mongo to use Fluentd's tag to determine the destination collection.
For example, if you generate records with tags mongo.foo, the records will be inserted into the foo collection within the fluentd database:
1
<match mongo.*>
2
  @type mongo_replset
3
  database fluentd
4
  nodes localhost:27017,localhost:27018,localhost:27019
5
​
6
  # Set 'tag_mapped' if you want to use tag mapped mode.
7
  tag_mapped
8
​
9
  # If the tag is "mongo.foo", then the prefix "mongo." is removed.
10
  # The inserted collection name is "foo".
11
  remove_tag_prefix mongo.
12
​
13
  # This configuration is used if the tag is not found. The default is 'untagged'.
14
  collection misc
15
</match>
Copied!
replica_set
type
default
version
string
required parameter
1.0.0
The name of the replica set.
read
type
default
version
string
nil
1.0.0
The ReplicaSet read preference (e.g. secondary, etc.).
num_retries
type
default
version
integer
60
1.0.0
The ReplicaSet failover threshold. The default threshold is 60. If the retry count reaches this threshold, the plugin raises an exception.
Common Output / Buffer parameters
For common output / buffer parameters, please check the following articles:
​Output Plugin Overview​
​Buffer Section Configuration​
Further Readings
​fluent-plugin-webhdfs mongo​
If this article is incorrect or outdated, or omits critical information, please let us know. Fluentd is an open-source project under Cloud Native Computing Foundation (CNCF). All components are available under the Apache 2 License.