Configuring Periodic MIB Data Collection and Transfer

This document describes how to periodically transfer selected MIB data from your router to a specified Network Management System (NMS). The periodic MIB data collection and transfer feature is also known as bulk statistics.

Prerequisites for Periodic MIB Data Collection and Transfer

To use periodic MIB data collection and transfer, you should be familiar with the Simple Network Management Protocol (SNMP) model of management information. You should also know what MIB information you want to monitor on your network devices, and the OIDs or object names for the MIB objects to be monitored.

Information About Periodic MIB Data Collection and Transfer

SNMP Objects and Instances

A type (or class) of SNMP management information is called an object. A specific instance from a type of management information is called an object instance (or SNMP variable). To configure a bulk statistics collection, you must specify the object types to be monitored using a bulk statistics object list and the specific instances of those objects to be collected using a bulk statistics schema.

MIBs, MIB tables, MIB objects, and object indices can all be specified using a series of numbers called an object identifier (OID). OIDs are used in configuring a bulk statistics collection in both the bulk statistics object lists (for general objects) and in the bulk statistics schemas (for specific object instances).

Bulk Statistics Object Lists

To group the MIB objects to be polled, you need to create one or more object lists. A bulk statistics object list is a user-specified set of MIB objects that share the same MIB index. Object lists are identified using a name that you specify. Named bulk statistics object lists allow the same configuration to be reused in different bulk statistics schemas.

All the objects in an object list must share the same MIB index. However, the objects do not need to be in the same MIB and do not need to belong to the same MIB table. For example, it is possible to group ifInOctets and a CISCO-IF-EXTENSION-MIB object in the same schema, because the containing tables for both objects are indexed by the ifIndex.

Bulk Statistics Schemas

Data selection for the Periodic MIB Data Collection and Transfer Mechanism requires the definition of a schema with the following information:

  • Name of an object list.

  • Instance (specific instance or series of instances defined using a wild card) that needs to be retrieved for objects in the specified object list.

  • How often the specified instances need to be sampled (polling interval). The default polling interval is 5 minutes.

A bulk statistics schema is also identified using a name that you specify. This name is used when configuring the transfer options.

Bulk Statistics Transfer Options

After configuring the data to be collected, a single virtual file (VFile or bulk statistics file) with all collected data is created. This file can be transferred to a network management station using FTP or TFTP. You can specify how often this file should be transferred. The default transfer interval is once every 30 minutes. You can also configure a secondary destination for the file to be used if, for whatever reason, the file cannot be transferred to the primary network management station.

The value of the transfer interval is also the collection period (collection interval) for the local bulk statistics file. After the collection period ends, the bulk statistics file is frozen, and a new local bulk statistics file is created for storing data. The frozen bulk statistics file is then transferred to the specified destination.

By default, the local bulk statistics file is deleted after successful transfer to an network management station.

Benefits of Periodic MIB Data Collection and Transfer

Periodic MIB data collection and transfer (bulk statistics feature) allows many of the same functions as the bulk file MIB (CISCO-BULK-FILE-MIB.my), but offers some key advantages. The main advantage is that this feature can be configured through the CLI and does not require an external monitoring application.

Periodic MIB data collection and transfer is mainly targeted for medium to high-end platforms that have sufficient local storage (volatile or permanent) to store bulk statistics files. Locally storing bulk statistics files helps minimize loss of data during temporary network outages.

This feature also has more powerful data selection features than the bulk file MIB; it allows grouping of MIB objects from different tables into data groups (object lists). It also incorporates a more flexible instance selection mechanism, where the application is not restricted to fetching an entire MIB table.

How to Configure Periodic MIB Data Collection and Transfer

Configuring a Bulk Statistics Object List

The first step in configuring the Periodic MIB Data Collection and Transfer Mechanism is to configure one or more object lists.

SUMMARY STEPS

  1. configure
  2. snmp-server mib bulkstat object-list list-name
  3. add {oid | object-name}
  4. Use the commit or end command.

DETAILED STEPS

  Command or Action Purpose
Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters global configuration mode.

Step 2

snmp-server mib bulkstat object-list list-name

Example:

snmp-server mib bulkstat object-list ifMib

Defines an SNMP bulk statistics object list and enters bulk statistics object list configuration mode.

Step 3

add {oid | object-name}

Example:


RP/0/RP0/CPU0:router(config-bulk-objects)# add 1.3.6.1.2.1.2.2.1.11
RP/0/RP0/CPU0:router(config-bulk-objects)# add ifAdminStatus
RP/0/RP0/CPU0:router(config-bulk-objects)# add ifDescr

Adds a MIB object to the bulk statistics object list. Repeat as desired until all objects to be monitored in this list are added.

Note 

All the objects in a bulk statistics object list have to be indexed by the same MIB index. However, the objects in the object list do not need to belong to the same MIB or MIB table.

When specifying an object name instead of an OID (using the add command), only object names with mappings shown in the show snmp mib object command output can be used.

Step 4

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Configuring a Bulk Statistics Schema

The second step in configuring periodic MIB data collection and transfer is to configure one or more schemas.

Before you begin

The bulk statistics object list to be used in the schema must be defined.

SUMMARY STEPS

  1. configure
  2. snmp-server mib bulkstat schema schema-name
  3. object-list list-name
  4. Do one of the following:
    • instance exact {interface interface-id [sub-if] | oid oid}
    • instance wild {interface interface-id [sub-if] | oid oid}
    • instance range start oid end oid
    • instance repetition oid max repeat-number
  5. poll-interval minutes
  6. Use the commit or end command.

DETAILED STEPS

  Command or Action Purpose
Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters global configuration mode.

Step 2

snmp-server mib bulkstat schema schema-name

Example:


RP/0/RP0/CPU0:router(config)# snmp-server mib 
bulkstat schema intE0
RP/0/RP0/CPU0:router(config-bulk-sc)#

Names the bulk statistics schema and enters bulk statistics schema mode.

Step 3

object-list list-name

Example:

RP/0/RP0/CPU0:router(config-bulk-sc)# object-list 
ifMib

Specifies the bulk statistics object list to be included in this schema. Specify only one object list per schema. If multiple object-list commands are executed, the earlier ones are overwritten by newer commands.

Step 4

Do one of the following:

  • instance exact {interface interface-id [sub-if] | oid oid}
  • instance wild {interface interface-id [sub-if] | oid oid}
  • instance range start oid end oid
  • instance repetition oid max repeat-number

Example:

RP/0/RP0/CPU0:router(config-bulk-sc)# instance 
wild oid 1
or 
RP/0/RP0/CPU0:router(config-bulk-sc)# instance 
exact interface TenGigE 0/1.25
or 
RP/0/RP0/CPU0:router(config-bulk-sc)# instance 
range start 1 end 2
or 
RP/0/RP0/CPU0:router(config-bulk-sc)# instance 
repetition 1 max 4

Specifies the instance information for objects in this schema:

  • The instance exact command indicates that the specified instance, when appended to the object list, represents the complete OID.

  • The instance wild command indicates that all subindices of the specified OID belong to this schema. The wild keyword allows you to specify a partial, “wild carded” instance.

  • The instance range command indicates a range of instances on which to collect data.

  • The instance repetition command indicates data collection to repeat for a certain number of instances of a MIB object.

Note 

Only one instance command can be configured per schema. If multiple instance commands are executed, the earlier ones are overwritten by new commands.

Step 5

poll-interval minutes

Example:

RP/0/RP0/CPU0:router(config-bulk-sc)# poll-interval 10

Sets how often data should be collected from the object instances specified in this schema, in minutes. The default is once every 5 minutes. The valid range is from 1 to 20000.

Step 6

Use the commit or end command.

commit —Saves the configuration changes and remains within the configuration session.

end —Prompts user to take one of these actions:

  • Yes — Saves configuration changes and exits the configuration session.

  • No —Exits the configuration session without committing the configuration changes.

  • Cancel —Remains in the configuration session, without committing the configuration changes.

Configuring Bulk Statistics Transfer Options

The final step in configuring periodic MIB data collection and transfer is to configure the transfer options. The collected MIB data are kept in a local file-like entity called a VFile (virtual file, referred to as a bulk statistics file in this document). This file can be transferred to a remote network management station at intervals you specify.

Before you begin

The bulk statistics object lists and bulk statistics schemas must be defined before configuring the bulk statistics transfer options.

SUMMARY STEPS

  1. configure
  2. snmp-server mib bulkstat transfer-id transfer-id
  3. buffer-size bytes
  4. format {bulkBinary | bulkASCII | schemaASCII}
  5. schema schema-name
  6. transfer-interval minutes
  7. url primary url
  8. url secondary url
  9. retry number
  10. retain minutes
  11. enable
  12. commit minutes

DETAILED STEPS

  Command or Action Purpose
Step 1

configure

Example:


RP/0/RP0/CPU0:router# configure

Enters global configuration mode.

Step 2

snmp-server mib bulkstat transfer-id transfer-id

Example:

RP/0/RP0/CPU0:router(config)# snmp-server mib 
bulkstat transfer bulkstat1

Identifies the transfer configuration with a name (transfer-id argument) and enters bulk statistics transfer configuration mode.

Step 3

buffer-size bytes

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# buffersize 3072

(Optional) Specifies the maximum size for the bulk statistics data file, in bytes. The valid range is from 1024 to 2147483647 bytes. The default buffer size is 2048 bytes.

Note 

If the maximum buffer size for a bulk statistics file is reached before the transfer interval time expires, all additional data received is deleted. To correct this behavior, you can decrease the polling frequency, or increase the size of the bulk statistics buffer.

Step 4

format {bulkBinary | bulkASCII | schemaASCII}

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# format schemaASCII

(Optional) Specifies the format of the bulk statistics data file (VFile). The default is schemaASCII.

Note 

Transfers can only be performed using schemaASCII (cdcSchemaASCII) format. SchemaASCII is a human-readable format that contains parser-friendly hints for parsing data values.

Step 5

schema schema-name

Example:


RP/0/RP0/CPU0:router(config-bulk-tr)# schema TenGigE 0/5/0/11/1
RP/0/RP0/CPU0:router(config-bulk-tr)# schema TenGigE/0-CAR
RP/0/RP0/CPU0:router(config-bulk-tr)# schema TenGigE 0/5/0/11/1

Specifies the bulk statistics schema to be transferred. Repeat this command as desired. Multiple schemas can be associated with a single transfer configuration; all collected data are placed in a single bulk data file (VFile).

Step 6

transfer-interval minutes

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# transfer-interval 20

(Optional) Specifies how often the bulk statistics file are transferred, in minutes. The default value is once every 30 minutes. The transfer interval is the same as the collection interval.

Step 7

url primary url

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# url primary 
ftp://user:password@host/folder/bulkstat1

Specifies the network management system (host) that the bulk statistics data file is transferred to, and the protocol to use for transfer. The destination is specified as a Uniform Resource Locator (URL). FTP or TFTP can be used for the bulk statistics file transfer.

Step 8

url secondary url

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# url secondary 
tftp://10.1.0.1/tftpboot/user/bulkstat1

(Optional) Specifies a backup transfer destination and protocol for use in the event that transfer to the primary location fails. FTP or TFTP can be used for the bulk statistics file transfer.

Step 9

retry number

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# retry 1

(Optional) Specifies the number of transmission retries. The default value is 0 (in other words, no retries). If an attempt to send the bulk statistics file fails, the system can be configured to attempt to send the file again using this command.

One retry includes an attempt first to the primary destination then, if the transmission fails, to the secondary location. For example, if the retry value is 1, an attempt is made first to the primary URL, then to the secondary URL, then to the primary URL again, then to the secondary URL again. The valid range is from 0 to 100.

If all retries fail, the next normal transfer occurs after the configured transfer-interval time.

Step 10

retain minutes

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# retain 60

(Optional) Specifies how long the bulk statistics file should be kept in system memory, in minutes, after the completion of the collection interval and a transmission attempt is made. The default value is 0. Zero (0) indicates that the file is deleted immediately after the transfer is attempted. The valid range is from 0 to 20000.

Note 

If the retry command is used, you should configure a retain interval larger than 0. The interval between retries is the retain interval divided by the retry number. For example, if retain 10 and retry 2 are configured, two retries are attempted once every 5 minutes. Therefore, if retain 0 is configured, no retries are attempted.

Step 11

enable

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# enable

Begins the bulk statistics data collection and transfer process for this configuration.

  • For successful execution of this action, at least one schema with non-zero number of objects must be configured.

  • Periodic collection and file transfer begins only if this command is configured. Conversely, the no enable command stops the collection process. A subsequent enable starts the operations again.

  • Each time the collection process is started using the enable command, data is collected into a new bulk statistics file. When the no enable command is used, the transfer process for any collected data immediately begins (in other words, the existing bulk statistics file is transferred to the specified management station).

Step 12

commit minutes

Example:

RP/0/RP0/CPU0:router(config-bulk-tr)# retain 60

If the maximum buffer size for a bulk statistics file is reached before the transfer interval time expires, the transfer operation is still initiated, but any bulk statistics data received after the file was full, and before it was transferred, are deleted. To correct this behavior, you can decrease the polling frequency, or increase the size of the bulk statistics buffer.

If retain 0 is configured, no retries are attempted. This is because the interval between retries is the retain value divided by the retry value. For example, if retain 10 and retry 2 are configured, retries are attempted once every 5 minutes. Therefore, if you configure the retry command, you should also configure an appropriate value for the retain command.

Periodic MIB Data Collection and Transfer: Example

This example shows how to configure periodic MIB data collection and transfer: 


snmp-server mib bulkstat object-list cempo
add cempMemPoolName
add cempMemPoolType
!
snmp-server mib bulkstat schema cempWild
object-list cempo
instance wild oid 8695772
poll-interval 1
!
snmp-server mib bulkstat schema cempRepeat
object-list cempo
instance repetition 8695772.1 max 4294967295
poll-interval 1
!
snmp-server mib bulkstat transfer-id cempt1
enable
url primary tftp://223.255.254.254/auto/tftp-sjc-users3/username/dumpdcm
schema cempWild
schema cempRepeat
transfer-interval 2
!

This example shows sample bulk statistics file content: 


Schema-def cempt1.cempWild "%u, %s, %s, %d" Epochtime instanceoid 
            1.3.6.1.4.1.9.9.221.1.1.1.1.3 1.3.6.1.4.1.9.9.221.1.1.1.1.2
cempt1.cempWild: 1339491515, 8695772.1, processor, 2
cempt1.cempWild: 1339491515, 8695772.2, reserved, 11
cempt1.cempWild: 1339491515, 8695772.3, image, 12
cempt1.cempWild: 1339491575, 8695772.1, processor, 2
cempt1.cempWild: 1339491575, 8695772.2, reserved, 11
cempt1.cempWild: 1339491575, 8695772.3, image, 12
Schema-def cempt1.cempRepeat "%u, %s, %s, %d" Epochtime instanceoid 
            1.3.6.1.4.1.9.9.221.1.1.1.1.3 1.3.6.1.4.1.9.9.221.1.1.1.1.2
cempt1.cempRepeat: 1339491515, 8695772.1, processor, 2
cempt1.cempRepeat: 1339491515, 8695772.2, reserved, 11
cempt1.cempRepeat: 1339491515, 8695772.3, image, 12
cempt1.cempRepeat: 1339491515, 26932192.1, processor, 2
cempt1.cempRepeat: 1339491515, 26932192.2, reserved, 11
cempt1.cempRepeat: 1339491515, 26932192.3, image, 12
cempt1.cempRepeat: 1339491515, 35271015.1, processor, 2
cempt1.cempRepeat: 1339491515, 35271015.2, reserved, 11
cempt1.cempRepeat: 1339491515, 35271015.3, image, 12
cempt1.cempRepeat: 1339491515, 36631989.1, processor, 2
cempt1.cempRepeat: 1339491515, 36631989.2, reserved, 11
cempt1.cempRepeat: 1339491515, 36631989.3, image, 12
cempt1.cempRepeat: 1339491515, 52690955.1, processor, 2
cempt1.cempRepeat: 1339491515, 52690955.2, reserved, 11
cempt1.cempRepeat: 1339491515, 52690955.3, image, 12