Chapter 1: Introduction

	cOS Stream 4.00.03 Statistics Reference Guide	Next

	Note: This document is also available in other formats
A PDF version of this document along with all current and older documentation in PDF format can be found at https://my.clavister.com. It is also available as a single HTML page.

Note: This document is also available in other formats

A PDF version of this document along with all current and older documentation in PDF format can be found at https://my.clavister.com.

It is also available as a single HTML page.

The Clavister cOS Stream maintains a set of statistical values to provide detailed runtime information for the administrator. These statistics provide information about both the current state of the system as well as information about the recent history.

The bulk of this guide lists the individual statistical values that are available. This chapter explains how the values are grouped and how to interpret the information presented for each statistical value.

The guide is divided into two major sections:

Clavister Statistics

This section lists the Clavister proprietary statistical values that are maintained by the Clavister cOS Stream. It includes both the statistical values that are accesible from the cOS Stream CLI as well as all statistical values that are included in the Clavister proprietary MIB. Some values available in the CLI are not part of the MIB, while other values available in the MIB are not accessible from the CLI.

The MIB files are supplied with the Clavister cOS Stream and have the filenames CLAVISTER-SMI.mib and CLAVISTER-STREAM.mib. These can be downloaded using SCP from the system's top-level directory.
Supported Standard MIBs

This section describes all standard supported MIB files and the statistical values from those MIBs that are supported. It is divided into one section per standard MIB that is at least partially supported. Subsections then list the statistical values from that MIB that are supported.

The supported standard MIBs files are not supplied by Clavister but are available from third parties. The statistical values listed are those available using an SNMP client and these may not include all the values in the MIB.

Statistics Parameters

For both the Clavister statistics section and every standard MIB, the following parameters are optionally listed for each available statistical value:

Name

The textual name of the statistical value.

Description

A textual description of the statistical value and what it represents.

Category

The data category for this statistical value. This can be one of:

Counter - The value is positive and increments.
Momentaneous - The value can vary arbitrarily.
Min - Only the the smallest observed value is stored.
Max - Only the the largest observed value is stored.

Type

The data type of the statistical value. For example, a counter will usually be an Unsigned 64-bit integer. The possibilities are:

8 bit unsigned integer - Range: 0 to 255.
16 bit unsigned integer - Range: 0 to 65,535.
32 bit unsigned integer - Range: 0 to 2^32-1.
64 bit unsigned integer - Range: 0 to 2^64-1.
8 bit signed integer - Range: -128 to 127.
16 bit signed integer - Range: -32,768 to 32,767.
32 bit signed integer - Range: -2^31 to 2^31-1.
64 bit signed integer - Range: -2^63 to 2^63-1.
String - Human readable string.
Byte array - Hexadecimal string.

Unit

Unit of measurement. For example, seconds or packets-per-second or bytes. The unit is often postfixed to the current value when it is displayed.

Stat Path

When present, this is the unique identifier used for accessing the statistic through the CLI. It is built in a similar way to a file path starting with the main context. This is possibly followed by the sub-context and finally with the identifier for this specific statistical value.

A Stat Path can take the following form:

		/ifaces/[..n]/bytes_in

The [..n] notation in the Stat Path indicates that this part will be consist of a dynamic name. In the above case, [..n] will be replaced with an actual interface name. There will be several interface names and these may vary from platform to platform. However, there will only be one bytes_in statistical value per interface.

If the name of the interface is If1 then the actual path name becomes:

		/ifaces/If1/bytes_in

Whenever the [..n] notation is seen in this reference guide, an actual value for that part of the path will have to be inserted.

MIB OID

The unique Object Identifier (OID) for this statistical value. The OID is used by SNMP clients to access a specific statistical value.

An MIB OID is built up from a sequence of numbers divided by periods ("."). The following examples show how it may be presented for a specific statistical value:

MIB OID 1.3.6.1.2.1.1.3[.0] (MIB-II, RFC1213)
MIB OID 1.3.6.1.4.1.5089.1.2.1.7.1.3[.n] (STREAM-MIB)
MIB OID 1.3.6.1.4.1.5089.3.2.2090.3020.1.2[.n.m] (STREAM-MIB)

The number sequence before the square brackets [ ] forms the OID of an object type, as specified in the MIB. This is also referred to as the OID prefix. The part within the square brackets [ ] is the instance sub-identifier, referred to as an OID fragment. It is appended to the OID prefix to form the complete and unique OID for an instance. Requests of the type SNMP GET always need a complete OID, including the instance sub-identifier. Requests of the type SNMP GETNEXT can use either just an OID prefix or a complete OID, depending on what the intent is. Successful SNMP response messages always contain complete OIDs to uniquely identify which instance's value was returned.

Any one of the following notations may occur within the square brackets:

The [.0] notation indicates that this is a single-instance object, where ".0" should be appended to the OID to identify that unique object instance.
The [.n] notation indicates that the object has a dynamic number of instances, where a dynamic number (index n) should be appended to the OID to identify a unique object instance. These instances can be seen as the rows in a table of values. In the example above, there will be one instance of this statistical value for each interface. This means that the index n is simply increased for each interface.
The [.n.m] notation also indicates that the object has a dynamic number of instances. The difference is that these instances are accessed using a two-dimensional instance identifier, representing a table within a table.

The text within the ending parentheses ( ) refers to the MIB, and optionally also the RFC, where this particular statistical value is defined. In the second and third examples above, CLAVISTER-STREAM.mib therefore needs to be loaded into the SNMP client to be able to fetch these particular statistical values.

MIB Name

This is the textual name for the statistical value, as it is specified in the MIB definition.

MIB Type

This is the type for this statistical value when accessed via SNMP with an SNMP client. The possible values are:

Counter - Range: 0 to 2^32-1.
Counter32 - Range: 0 to 2^32-1.
Counter64 - Range: 0 to 2^64-1.
CounterBasedGauge64 - Range: 0 to 2^64-1.
DisplayString - Human readable string.
Gauge32 - Range: 0 to 2^31-1.
Integer - Range: -2^31 to 2^31-1.
Integer32 - Range: -2^31 to 2^31-1.
IpAddress - IPv4 address represented as an octet string of length 4.
ObjectIdentifier - String of dot separated digits.
OCTET STRING - Hexadecimal string representing arbitrary data.
PhysAddress - Hexadecimal string.
TimeTicks - Hundreds of a second, 0 to 2^32.
Unsigned32 - Range: 0 to 2^32-1.

The MIB types are described further in the relevant RFC documents.

2nd OID

When a statistical value is made available in more than one MIB, and therefore listed in more than one place in this document, this specifies the OID in the additional MIB and which MIB it is a part of.

2nd Name

The name associated with the 2nd OID.

2nd Type

The type associated with the 2nd OID.