software.amazon.jdbc.plugin.efm

Class HostMonitorConnectionContext

java.lang.Object

software.amazon.jdbc.plugin.efm.HostMonitorConnectionContext

public class HostMonitorConnectionContext
extends java.lang.Object

Monitoring context for each connection. This contains each connection's criteria for whether a server should be considered unhealthy. The context is shared between the main thread and the monitor thread.

Constructor Summary

Constructors

Constructor and Description
HostMonitorConnectionContext(HostMonitor monitor, java.sql.Connection connectionToAbort, long failureDetectionTimeMillis, long failureDetectionIntervalMillis, long failureDetectionCount, TelemetryCounter abortedConnectionsCounter) Constructor.

Method Summary

Modifier and Type	Method and Description
long	getExpectedActiveMonitoringStartTimeNano()
long	getFailureCount()
long	getFailureDetectionCount()
long	getFailureDetectionIntervalMillis()
java.util.concurrent.locks.ReentrantLock	getLock()
HostMonitor	getMonitor()
boolean	isActiveContext()
boolean	isNodeUnhealthy()
void	setInactive()
void	updateConnectionStatus(java.lang.String hostName, long statusCheckStartTimeNano, long statusCheckEndTimeNano, boolean isValid) Update whether the connection is still valid if the total elapsed time has passed the grace period.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

HostMonitorConnectionContext

public HostMonitorConnectionContext(HostMonitor monitor,
                                    java.sql.Connection connectionToAbort,
                                    long failureDetectionTimeMillis,
                                    long failureDetectionIntervalMillis,
                                    long failureDetectionCount,
                                    TelemetryCounter abortedConnectionsCounter)

Constructor.

Parameters:

monitor - A reference to a monitor object.

connectionToAbort - A reference to the connection associated with this context that will be aborted in case of server failure.

failureDetectionTimeMillis - Grace period after which node monitoring starts.

failureDetectionIntervalMillis - Interval between each failed connection check.

failureDetectionCount - Number of failed connection checks before considering database node as unhealthy.

abortedConnectionsCounter - Aborted connection telemetry counter.

Method Detail

getFailureDetectionIntervalMillis

public long getFailureDetectionIntervalMillis()

getFailureDetectionCount

public long getFailureDetectionCount()

getFailureCount

public long getFailureCount()

getExpectedActiveMonitoringStartTimeNano

public long getExpectedActiveMonitoringStartTimeNano()

getMonitor

public HostMonitor getMonitor()

isNodeUnhealthy

public boolean isNodeUnhealthy()

isActiveContext

public boolean isActiveContext()

setInactive

public void setInactive()

updateConnectionStatus

public void updateConnectionStatus(java.lang.String hostName,
                                   long statusCheckStartTimeNano,
                                   long statusCheckEndTimeNano,
                                   boolean isValid)

Update whether the connection is still valid if the total elapsed time has passed the grace period.

Parameters:

hostName - A node name for logging purposes.

statusCheckStartTimeNano - The time when connection status check started in nanos.

statusCheckEndTimeNano - The time when connection status check ended in nanos.

isValid - Whether the connection is valid.

getLock

public java.util.concurrent.locks.ReentrantLock getLock()