• Major Software functions: This interface embodies section 1.3 of the Distributed Network Manager Functional Specification. The Network Management Application shall interact with and take advantage of existing SNMP resources to collect information. This module will support the SNMP MIB II data structure. This module will also be developed in full compliance with the ISO Abstract Syntax Notation (ASN.1) used to specify SNMP MIBs. This version will support SNMP v1.

• System Constraints: The SNMP manager is the only requirement addressed by this interface. It does not include features such as node-discovery, security, or statistical analysis. This implementation does not support outside of the MIB II data structure although, can be extended in the future. There is not interface defined between the Privileged User Interface or Node Discovery, instead a staitic configuration file is being used as input. This module can easily be extended to accommodate such an interface when it is defined. This module does not monitor resource utilization such as CPU load.

• References: This design has been derived from v1.0 of the Distributed Network Manager Functional Specification(Oliver), the Software Requirements Specification v1.0(Waiters) and the Software Design Document v1.1(Howard). The SNMP standards implemented comply with SNMP as defined in RFCs 1115, 1156, 1157, 1158, 1212 and 1213.

• Data Flow Overview:
  Level 0
  (context model)
  

  Level 1
  (SNMP Manager)

  

• Interfaces: There are three modules for this interface. Each module contains one function. The scope of these three modules is to fill in the "Entry" data structure defined below. The modules are as follows:
  • Parser, parser is passed a pointer to the query structure.

  • SNMP_get, this function is given a pointer the query and the entry structure.

  • Output, receives a pointer to the entry structure and outputs to the report file.

• Data Structures: There are two main data structures for this module. The parser and SNMP_get functions deal with the query data structure. SNMP_get and the output functions work with the entry data structure. They are represented as shown.

  Query Data Structure
  Name	Type
  node	String
  community	String
  object _ID	String
  next	pointer (query)

  Entry Data Structure
  Name	Type
  node	String
  timestamp	time
  field_value	long integer
  next	pointer (entry)

  Time Data Structure
  Name	Type
  second	integer
  minute	integer
  hour	integer
  day	integer
  year 1	(19-99) integer
  year 2	(0-99) integer

• The parser opens the ASCII configuration file containing the node, community , and MIB object Ids and parses them into the query data structure which gets passed to SNMP_get. If there is an error in the configuration file a CONFIG_ERROR exception is raised. Note: community is used for SNMP conformance only and should not be considered secure.

• The SNMP_get function uses the query data structure to form the SNMP get requests. The query data structure is a linked list to provide for more than one SNMP get request for a particular node. Each SNMP get is time stamped and stored in the entry data structure. If request is not returned by the SNMP agent the field_value_entry is set to the string constant "UNACKNOWLEDGED".

• The output function receives the entry data structure which is also a linked list and iterates through the list to produce the output file. If the file is locked (i.e. is being processed or modified by the client) the function will enter a random time-out/retry loop until the file can be written to.

• Configuration File contains entries for SNMP gets. This file is accessed as read-only by the parser function. There are three fields separated by spaces or tabs. Each entry is delimited by a carriage return and line-feed. These fields are as follows;

  Field name	Data Type	Comments
  Node	String	DNS registered name
  Community	String
  Object ID(s)	String	Text form only
  Period	Integer	Seconds

• Report File contains the entries needed for statistical analysis. The report file is created in a write/append mode. The client of this file is responsible for deleting removing unneeded information from this file. This format was provided by the statistical analysis groups as follows. There are three fields separated by spaces or tabs. Each entry is delimited by a carriage return and line-feed. These fields are as follows;

  Field name	Data Type	Comments
  Node	String
  Time stamp	String
  field entry value(s)	String

Design Item	Requirements Specification
Data Representation & Organization	0.3 Functional(v1.0) & 13. SDD(v1.1)
SNMP Manager/Agent Interface	1.3 Functional(v1.0) & 2.2 Software(v1.0)
Protocol Module(s)	2 Functional(v1.0) & 2.2 Software(v1.0)
Simple Data Collection	4.1 Functional(v1.0) & 2.2 Software(v1.0)