[Erlang Systems]

1 Introduction

The operation and maintenance support in OTP consists of a generic model for management subsystems in OTP, and some components to be used in these subsystems. The model that this support is based upon is described in "OAM Principles".

This document describes one of these components, the Fault Management application EVA. EVA consists of support for Event and Alarm Handling and support for generic Log Control.

EVA is a modular application that consists of two management protocol independent services. It contains also rules and functions for defining protocol adaptations for EVA. One such adaptation is included for SNMP. It consists of SNMP MIBs and implementation of these MIBs.

EVA uses the applications Mnesia and SASL.

1.1 Architecture

EVA is a subapplication that can be included into another application. It is designed to work as a distributed application, which means that it always executes on one node, with other nodes as standby nodes. EVA should run on the same node as other operation and maintenance applications, specifically the management protocol termination application, to minimize internal network traffic.

EVA is designed to be protocol independent, and may be used with different management protocols. For each such protocol, an EVA adaptation must be written. For example, we have defined an SNMP adaptation, and more adaptations may be defined in the future, e.g. for HTTP, CORBA or any proprietary protocol (e.g. plain Erlang).

The event and alarm support can run in two different modes, in server mode and in client mode. In client mode, no processes are running, but the code implementing the API is loaded. There must always be one server running on one node in the network of Erlang nodes that the system consists of.

For EVA, all involved nodes are seen as one (distributed) system. This means for example that there is one active alarm list for the entire system.


Copyright © 1991-2001 Ericsson Utvecklings AB