Skip to content. Skip to navigation

What Clients Are saying

Document Actions
course id
SNMP-AGENT
duration
1 day(s)
Course Title SNMP: Agent Development
Aimed At

SNMP Agent software developers.

Course in a Nutshell

After an Enterprise’s MIB design is finalized and fully validated, the next task is Agent software development. A number of issues need to be considered in selecting a development kit.

  • Targeted operating system (or systems)
  • Pre-existence of SNMP agents on those platforms
  • Embedded platforms or not
  • Tool cost vs. time-to-market (and market opportunity cost)
  • Protocol validation issues

Once a tool is selected, many common coding issues will be encountered. This intensive, one-day tutorial will prepare you to deal with these issues and allow you to proceed quickly to a deliverable product. Example topics include choosing how to best access and manage MIB-related data, and the number (and purposes) of agent code threads that may be needed. We will reveal a number of very surprising problems drawn from our own extensive SNMP experience related to deployed agents, followed by a discussion of agent testing strategies, descriptions of specific tests, and the presentation of a comprehensive test plan from a past project. You will learn the important details of syntax encoding for “get”, “set”, “trap” and other message types. The course will conclude with the review of a checklist for productizing Enterprise MIBs and agents.

Customize It!
Learn How To
  • Watch out for common agent problems
  • Make agent architectural and SDK choices
  • Define agent semantics
  • Coexist (if necessary) with native agents
  • Be aware of important coding issues
  • Describe the details of SNMP message syntax
  • Fully test your agent prior to deployment
  • Productize your MIB and agent products
Course Outline
  • True stories of delivered agents with significant deficiencies
  • Agent development steps
  • Defining goals and tradeoffs
  • Agent SDK selection criteria
  • Agent architectural choices and issues
  • Agent semantics
  • Coexistence of enterprise agents with native platform agents
  • Trap design
  • Agent coding issues
    • Data acquisition
    • Read-write data management
    • Returning multiple values
    • Processing get-next requests
    • Table index data type choices
    • Multi-step “set” processing
    • Code threads and uses
    • Trap send logic
    • Agent initialization and log files
  • SNMP message syntax and coding
  • Agent testing (static and dynamic)
  • Productizing an Enterprise MIB and agent
  • Wrap-up: Course recap, Q/A, and evaluations
Navigation