SNMP

• Source file config
• Pipeline file config
• Raw pipeline file config

Source file config

Property	Type	Description
type	String	Specify source type. Value - snmp
name	String	Unique source name - also the config filename
config	Object	Source configuration

All properties are required

config object properties:

Property	Type	Required	Description
hosts	Array	Yes	List of SNMP device URLs (default PORT is 161). The number of hosts for a single source shouldn't be more than 10
read_community	String	Yes	SNMP read community
query_timeout	Integer	No	Timeout value for SNMP query

Example

[
  {
    "type": "snmp",
    "name": "snmp",
    "config": {
      "hosts": ["http://somehost:161"],
      "read_community": "public",
      "query_timeout": 30
    }
  }
]

Pipeline File config

Properties list

Property	Required	Value type	Description
source	yes	String	Source config name
pipeline_id	yes	String	Unique pipeline identifier (use a human-readable name so you could easily use it further)
interval	yes	Integer	How often to query data, unit - seconds
dimension_value_paths	yes	Object	Key-value pairs of OIDs with dimension names that will be used as dimensions (dimension_name:oid)
oid_table	no	Object	Properties to process OIDs defined inside SNMP tables
values	yes	Object	Key-value pairs (target_type). A target type represents how samples of the same metric are aggregated in Anodot. Valid values are gauge (average aggregation), counter (sum aggregation), running_counter (will calculate the difference between the last value and current value)
units	no	Object	Key-value pairs (value:unit). The value must be from the values column, units can be any.
measurement_names	yes	Object	Key-value pairs (oid:name)
dimensions	no	Array	List of OIDs that will be used as dimensions
properties	no	Object	Dimensions with static values to pass to Anodot. Object with key-value pairs
tags	no	Object	Tags. Object with key-value pairs
transform	no	Object	See transformations page
notifications	No	Object	See notifications page

oid_table object properties:

Property	Required	Type	Description
<name>	yes	Object	SNMP Table name. Could be any
mib	yes	String	MIB Name. Without it is not possible to search dimensions or values by their names
use_indexes	no	Array	List of table index numbers to request. If empty - whole table requested
values	no	Object	Key-value pairs of OID names and (target_type). A target type represents how samples of the same metric are aggregated in Anodot. Valid values are gauge (average aggregation), counter (sum aggregation), running_counter (will calculate the difference between the last value and current value)
dimensions	no	Array	List of OIDs that will be used as dimensions.

Example

[
  {
    "source": "snmp",
    "pipeline_id": "snmp",
    "oid_table": {
      "ifTable": {
        "mib": "IF-MIB",
        "use_indexes": [2],
        "values": {
          "ifMtu": "counter",
          "ifSpeed": "gauge"
        },
        "dimensions": ["ifDescr", "ifType"]
      }
    },
    "values": {
      "1.3.6.1.2.1.5.14.0": "counter",
      "1.3.6.1.2.1.5.1.0": "counter"
    },
    "units": {
      "1.3.6.1.2.1.5.14.0": "unit_1",
      "1.3.6.1.2.1.5.1.0": "unit_1"
    },
    "measurement_names": {
      "1.3.6.1.2.1.5.14.0": "icmp_out_msgs",
      "1.3.6.1.2.1.5.1.0": "icmp_in_msgs"
    },
    "dimensions": ["system_description", "system_location"],
    "dimension_value_paths": {
      "system_description": "1.3.6.1.2.1.1.1.0",
      "system_location": "1.3.6.1.2.1.1.6.0",
      "hostname": "1.3.6.1.2.1.1.5.0",
    },
    "transform": {
      "config": "dimensions/host_name,zeus.snmplabs.com,'dimensions/host_name' == 'value=zeus.snmplabs.com (you can change this!)'"
    },
    "interval": "5m"
  }
]

Raw pipeline file config

Raw pipelines are designed specifically for test purposes only. Raw pipelines could be configured before standard pipeline usage. With this kind of pipelines, You can see what data is returned by the request to the data source, meaning what data contains in specified OIDs. It does not make any changes or transformations with data or send data to Anodot

To create a raw pipeline use the command agent pipeline create-raw -f path/to/config.json

Properties list

Property	Required	Value type in config file	Description
source	yes	String	Source config name
pipeline_id	yes	String	Unique pipeline identifier
interval	yes	Integer	Query data every N seconds
oids	yes	List	List of OIDS to retrieve from a device
send_data_format	no	String	Choose a data format in which the data will be saved to the filesystem. Possible values are: JSON, DELIMITED. Default: DELIMITED

Example:

[
  {
    "source": "snmp_raw",
    "pipeline_id": "snmp_raw",
    "oids": [
      "1.3.6.1.2.1.1.1.0",
      "1.3.6.1.2.1.1.2.0",
      "1.3.6.1.2.1.1.4.0"
    ],
    "interval": "5m"
  }
]