IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
« Rabbitmq Integration Plugin Input plugins »
Elastic Docs ›Logstash Reference [8.14] ›Integration plugins

SNMP Integration Plugin

edit
IMPORTANT: This documentation is no longer updated. Refer to Elastic's version policy and the latest documentation.

SNMP Integration Plugin

edit
  • Plugin version: v4.0.0
  • Released on: 2024-05-03
  • Changelog

This functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.

Technical Preview

The new integration-snmp plugin, and its component plugins--input-snmp and input-snmptrap--are available in Technical Preview and can be installed on the latest Logstash 7.x and 8.x versions.

Current 1.x versions of the input-snmp plugin are bundled with Logstash by default, and will soon be replaced by the snmp input plugin contained in this integration. (If you want to opt into the Technical Preview for the integration-snmp plugin, run bin/logstash-plugin install logstash-integration-snmp.)

Be aware of behavioral and mapping differences between current stand-alone plugins and the new versions included in the integration-snmp. The information in this topic can help.

Description

edit

The SNMP integration plugin includes:

  • SNMP input plugin
  • SNMPtrap input plugin

The new logstash-integration-snmp plugin combines the logstash-input-snmp and logstash-input-snmptrap plugins into a single Ruby gem. This integrated plugin package provides better alignment in SNMP processing behavior, improved resource management, easier package maintenance, and a smaller installation footprint. The individual plugins now share the same code base and have been refactored to leverage the latest version of SNMP4j.

Migrating to logstash-integration-snmp from individual plugins

edit

You’ll retain the functionality of existing stand-alone plugins, but with a more consistent experience and expanded features. In this section, we’ll note mapping and behavioral changes, and explain how to preserve current behavior if needed.

Migration notes: logstash-input-snmp

edit

As a component of the new logstash-integration-snmp plugin, the logstash-input-snmp plugin offers the same capabilities as the stand-alone logstash-input-snmp.

Given the underlying code refactor to better align snmp and snmptrap plugin behavior, you might need to address some changes depending on your use case and how the ingested data is being handled through the pipeline.

Changes to mapping and error logging: logstash-input-snmp
edit
  • No such instance errors are mapped as error: no such instance currently exists at this OID string instead of noSuchInstance.
  • No such object errors are mapped as error: no such object currently exists at this OID string instead of noSuchObject.
  • End of MIB view errors are mapped as error: end of MIB view instead of endOfMibView.
  • An unknown variable type falls back to the string representation instead of logging an error as it did in with the stand-alone logstash-input-snmp. This change should not affect existing pipelines, unless they have custom error handlers that rely on specific error messages.

Migration notes: logstash-input-snmptrap

edit

As a component of the new logstash-integration-snmp plugin, the logstash-input-snmptrap plugin offers almost the same capabilities as the stand-alone logstash-input-snmp plugin.

You might need to address some behavior changes depending on your use case and how the ingested data is being handled through the pipeline.

Changes to mapping and error logging: logstash-input-snmptrap
edit
  • The PDU variable bindings are mapped into the Logstash event using the defined data type. By default, the stand-alone logstash-input-snmptrap plugin converts all of the data to string, ignoring the original type. If this behavior is not what you want, you can use a filter to retain the original type.
  • SNMP TimeTicks variables are mapped as Long timestamps instead of formatted date string (%d days, %02d:%02d:%02d.%02d).
  • null variables values are mapped using the string null instead of Null (upper-case N).
  • No such instance errors are mapped as error: no such instance currently exists at this OID string instead of noSuchInstance.
  • No such object errors are mapped as error: no such object currently exists at this OID string instead of noSuchObject.
  • End of MIB view errors are mapped as error: end of MIB view instead of endOfMibView.

  • The previous generation (stand-alone) input-snmptrap plugin formatted the message field as a ruby-snmp SNMP::SNMPv1_Trap object representation.

    <SNMP::SNMPv1_Trap:0x6f1a7a4 @varbind_list=[#<SNMP::VarBind:0x2d7bcd8f @value="teststring", @name=[1.11.12.13.14.15]>], @timestamp=#<SNMP::TimeTicks:0x1af47e9d @value=55>, @generic_trap=6,  @enterprise=[1.2.3.4.5.6], @source_ip="127.0.0.1", @agent_addr=#<SNMP::IpAddress:0x29a4833e @value="test">, @specific_trap=99>

    The new integrated input-snmptrap plugin uses JSON to format message field.

    {"error_index":0, "variable_bindings":{"1.3.6.1.6.3.1.1.4.1.0":"SNMPv2-MIB::coldStart", "1.3.6.1.2.1.1.3.0":0}, "error_status":0, "type":"TRAP", "error_status_text":"Success", "community":"public", "version":"2c", "request_id":1436216872}
Maintain maximum compatibility with previous implementation
edit

If needed, you can configure the new logstash-integration-snmp plugin to maintain maximum compatibility with the previous (stand-alone) version of the input-snmp plugin.

input {
   snmptrap {
    use_provided_mibs => false
    oid_mapping_format => 'ruby_snmp'
    oid_map_field_values => true
   }
}

Importing MIBs

edit

The SNMP plugins already include the IETF MIBs (management information bases) and these do not need to be imported. To disable the bundled MIBs set the use_provided_mibs option to false.

Any other MIB will need to be manually imported to provide mapping of the numeric OIDs to MIB field names in the resulting event.

To import a MIB, the OSS libsmi library is required. libsmi is available and installable on most operating systems.

To import a MIB, you need to first convert the ASN.1 MIB file into a .dic file using the libsmi smidump command line utility.

Example (using RFC1213-MIB file)

$ smidump --level=1 -k -f python RFC1213-MIB > RFC1213-MIB.dic

Note that the resulting file as output by smidump must have the .dic extension.

Preventing a failed to locate MIB module error

edit

The smidump function looks for MIB dependencies in its pre-configured paths list. To avoid the failed to locate MIB module error, you may need to provide the MIBs locations in your particular environment.

The recommended ways to provide the additional path configuration are:

  • an environment variable, or
  • a config file to provide the additional path configuration.

See the "MODULE LOCATIONS" section of the smi_config documentation for more information.

Option 1: Use an environment variable

edit

Set the SMIPATH env var with the path to your MIBs. Be sure to include a prepended colon (:) for the path.

$ SMIPATH=":/path/to/mibs/" smidump -k -f python CISCO-PROCESS-MIB.mib > CISCO-PROCESS-MIB_my.dic

Notice the colon that precedes the path definition.

Option 2: Provide a configuration file

edit

The other approach is to create a configuration file with the path option. For example, you could create a file called smi.conf:

path :/path/to/mibs/

And use the config with smidump:

$ smidump -c smi.conf -k -f python CISCO-PROCESS-MIB.mib > CISCO-PROCESS-MIB_my.dic
« Rabbitmq Integration Plugin Input plugins »