servicemonitor.java

opennms得相关源码 请大家看看

//
// This file is part of the OpenNMS(R) Application.
//
// OpenNMS(R) is Copyright (C) 2002-2003 The OpenNMS Group, Inc.  All rights reserved.
// OpenNMS(R) is a derivative work, containing both original code, included code and modified
// code that was published under the GNU General Public License. Copyrights for modified 
// and included code are below.
//
// OpenNMS(R) is a registered trademark of The OpenNMS Group, Inc.
//
// Modifications:
//
// 2003 Jan 31: Added the ability to imbed RRA information in poller packages.
//
// Original code base Copyright (C) 1999-2001 Oculan Corp.  All rights reserved.
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 2 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
//
// For more information contact:
//      OpenNMS Licensing       <[e-mail]>
//      http://www.opennms.org/
//      http://www.opennms.com/
//

package org.opennms.netmgt.poller.monitors;

import java.util.Map;

import org.opennms.netmgt.config.PollerConfig;

/**
 * <p>
 * This is the interface that must be implemented by each poller plugin in the
 * framework. This well defined interface allows the framework to treat each
 * plugin identically.
 * </p>
 * 
 * <p>
 * When a service monitor plug-in is loaded and initialized, the framework will
 * initialize the monitor by calling the <EM>initialize()</EM> method.
 * Likewise, when the monitor is unloaded the framework calls the <EM>release()
 * </EM> method is called. If the plug-in needs to save or read any
 * configuration information after the initialize() call, a reference to the
 * proxy object should be saved at initialization.
 * </p>
 * 
 * <P>
 * <STRONG>NOTE: </STRONG> The plug-in <EM>poll()</EM> must be thread safe in
 * order to operate. Any synchronized methods or data accessed in the <EM>
 * poll()</EM> can negatively affect the framework if multiple poller threads
 * are blocked on a critical resource. Synchronization issues should be
 * seriously evaluated to ensure that the plug-in scales well to large
 * deployments.
 * </P>
 * 
 * @author <A HREF="mailto:[e-mail]">Brian Weaver </A>
 * @author <A HREF="http://www.opennms.org/">OpenNMS </A>
 * 
 */
public interface ServiceMonitor {
    /**
     * The constant the defines a status is unknown. Used mostly internally
     */
    public static final int SERVICE_UNKNOWN = 0;

    /**
     * <P>
     * The constant that defines a service as being in a normal state. If this
     * is returned by the poll() method then the framework will re-schedule the
     * service for its next poll using the standard uptime interval
     * </P>
     */
    public static final int SERVICE_AVAILABLE = 1;

    /**
     * <P>
     * The constant that defines a service that is not working normally and
     * should be scheduled using the downtime models.
     * </P>
     */
    public static final int SERVICE_UNAVAILABLE = 2;

    /**
     * <P>
     * The constant that defines a service that is up but is most likely
     * suffering due to excessive load or latency issues and because of that has
     * not responded within the configured timeout period.
     * </P>
     */
    public static final int SERVICE_UNRESPONSIVE = 3;

    /**
     * <P>
     * The status mask is used to mask off the bits that apply to scheduling
     * information only. The other bits of the status return information are
     * used to encode special processing instructions to the poller framework.
     * </P>
     */
    public static final int SERVICE_STATUS_MASK = 0xff;

    /**
     * <P>
     * This constants defines the mask that can be bitwise anded with the status
     * return from the poll() method to determine if a status event should be
     * surpressed. By default the framework will generate an event when the
     * status transitions from up to down or vice versa. If the mask bit is set
     * then the framework should not generate a transitional event.
     * </P>
     */
    public static final int SURPRESS_EVENT_MASK = 0x100;

    /**
     * <P>
     * This method is called after the framework creates an instance of the
     * plug-in. The framework passes the object a proxy object that can be used
     * to retreive configuration information specific to the plug-in.
     * Additionally, any parameters for the plug-in from the package definition
     * are passed using the parameters element.
     * </P>
     * 
     * <P>
     * If there is a critical error, like missing service libraries, the the
     * montior may throw a ServiceMonitorException. If the plug-in throws an
     * exception then the plug-in will be disabled in the framework.
     * </P>
     * 
     * @param parameters
     *            Not currently used
     * 
     * @exception java.lang.RuntimeException
     *                Thrown if an unrecoverable error occurs that prevents the
     *                plug-in from functioning.
     * 
     */
    public void initialize(PollerConfig config, Map parameters);

    /**
     * <P>
     * This method is called whenever the plug-in is being unloaded, normally
     * during framework exit. During this time the framework may release any
     * resource and save any state information using the proxy object from the
     * initialization routine.
     * </P>
     * 
     * <P>
     * Even if the plug-in throws a monitor exception, it will not prevent the
     * plug-in from being unloaded. The plug-in should not return until all of
     * its state information is saved. Once the plug-in returns from this call
     * its configuration proxy object is considered invalid.
     * </P>
     * 
     * @exception java.lang.RuntimeException
     *                Thrown if an error occurs during deallocation.
     * 
     */
    public void release();

    /**
     * <P>
     * This method is called whenever a new interface that supports the plug-in
     * service is added to the scheduling system. The plug-in has the option to
     * load and/or associate configuration information with the interface before
     * the framework begins scheduling the new device.
     * </P>
     * 
     * <P>
     * Should a monitor exception be thrown during an initialization call then
     * the framework will log an error and discard the interface from
     * scheduling.
     * </P>
     * 
     * @param iface
     *            The network interface to be added to the scheduler.
     * 
     * @exception java.lang.RuntimeException
     *                Thrown if an unrecoverable error occurs that prevents the
     *                interface from being monitored.
     * @exception
     *                org.opennms.netmgt.poller.monitors.NetworkInterfaceNotSupportedException
     *                Thrown if the passed interface is invalid for this
     *                monitor.
     * 
     */
    public void initialize(NetworkInterface iface);

    /**
     * <P>
     * This method is the called whenever an interface is being removed from the
     * scheduler. For example, if a service is determined as being no longer
     * supported then this method will be invoked to cleanup any information
     * associated with this device. This gives the implementor of the interface
     * the ability to serialize any data prior to the interface being discarded.
     * </P>
     * 
     * <P>
     * If an exception is thrown during the release the exception will be
     * logged, but the interface will still be discarded for garbage collection.
     * </P>
     * 
     * @param iface
     *            The network interface that was being monitored.
     * 
     * @exception java.lang.RuntimeException
     *                Thrown if an unrecoverable error occurs that prevents the
     *                interface from being monitored.
     */
    public void release(NetworkInterface iface);

    /**
     * <P>
     * This method is the heart of the plug-in monitor. Each time an interface
     * requires a check to be performed as defined by the scheduler the poll
     * method is invoked. The poll is passed the interface to check
     * </P>
     * 
     * <P>
     * By default when the status transition from up to down or vice versa the
     * framework will generate an event. Additionally, if the polling interval
     * changes due to an extended unavailbility, the framework will generate an
     * additional down event. The plug-in can surpress the generation of the
     * default events by setting the surpress event bit in the returned integer.
     * </P>
     * 
     * @param iface
     *            The network interface to test the service on.
     * @param parameters
     *            The package parameters (timeout, retry, etc...) to be used for
     *            this poll.
     * 
     * @return The availibility of the interface and if a transition event
     *         should be surpressed.
     * 
     * @exception java.lang.RuntimeException
     *                Thrown if an unrecoverable error occurs that prevents the
     *                interface from being monitored.
     * 
     * @see #SURPRESS_EVENT_MASK
     * @see #SERVICE_AVAILABLE
     * @see #SERVICE_UNAVAILABLE
     */
    public int poll(NetworkInterface iface, Map parameters, org.opennms.netmgt.config.poller.Package pkg);
}