Frank:
Thanks for your quick response.
Do you see any drawback from my code change?
In my application, the snmp agent thread co-exists with other threads.
By yielding to other threads when there is no message on the agent port,
other threads get more CPU time. I noticed that the overall CPU usage of
my application cuts into less than half comparing with the application
without this change.
I cut|paste my change here.
Have a nice day.
Vivi
=======================================================================
DatagramPacket packet = new DatagramPacket(buf, buf.length,
udpAddress.getInetAddress(),
udpAddress.getPort());
while (!stop) {
try {
try {
socket.receive(packet);
}
catch (InterruptedIOException iiox) {
if (iiox.bytesTransferred <= 0) {
try
{
Thread.sleep(getSocketTimeout() / 10);
}
catch (InterruptedException e)
{
// Now move on.
continue;
}
continue;
}
}
...
[email protected] wrote:
Send SNMP4J mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.agentpp.org/mailman/listinfo/snmp4j
or, via email, send a message with subject or body 'help' to
[email protected]
You can reach the person managing the list at
[email protected]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of SNMP4J digest..."
Today's Topics:
1. DefaultUdpTransportMapping class and ListenThread (Vivi Zhang)
2. Re: DefaultUdpTransportMapping class and ListenThread
(Frank Fock)
----------------------------------------------------------------------
Message: 1
Date: Mon, 02 Feb 2009 13:52:11 -0600
From: Vivi Zhang <[email protected]>
Subject: [SNMP4J] DefaultUdpTransportMapping class and ListenThread
To: [email protected].
Message-ID: <[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi,
I am looking at ways to tuning my application with embedded SNMP agent
with SNMP4J and SNMP4J-Agent.
I have two questions about the implementation of inner class ListernThread.
Q1) Line 315
* try {
socket.receive(packet);
}
catch (InterruptedIOException iiox) {
if (iiox.bytesTransferred <= 0) {
continue;
}*
If we see SocketTimeoutException since there is no packet to read,
can we put the thread to a small break before it goes for another
socket.receive() call?
The suggestion is intended to ease some pressure on the CPU and
network resource. Do you see any potential drawback from this change?
Q2) Line 307 *DatagramPacket packet = new DatagramPacket(buf, buf.length,
udpAddress.getInetAddress(),
udpAddress.getPort());**
*A DatagramPacket object is created repeatedly within the while
loop. What is the drawback if we move the stmt outiside while loop. The
logic behind this suggestion is avoiding object creating repeatedly
withing the loop.
Thanks.
Vivi
P.S.
I am using SNMP4J 1.9.3c.
Following is a section of code I cut from DefaultUdpTransportMapping.java.
============================================================================
while (!stop) {
* DatagramPacket packet = new DatagramPacket(buf, buf.length,
udpAddress.getInetAddress(),
udpAddress.getPort());*
try {
try {
socket.receive(packet);
}
catch (InterruptedIOException iiox) {
if (iiox.bytesTransferred <= 0) {
* continue;*
}
}
if (logger.isDebugEnabled()) {
logger.debug("Received message from "+packet.getAddress()+"/"+
packet.getPort()+
" with length "+packet.getLength()+": "+
new OctetString(packet.getData(), 0,
packet.getLength()).toHexString());
}
ByteBuffer bis;
// If messages are processed asynchronously (i.e. multi-threaded)
// then we have to copy the buffer's content here!
if (isAsyncMsgProcessingSupported()) {
byte[] bytes = new byte[packet.getLength()];
System.arraycopy(packet.getData(), 0, bytes, 0, bytes.length);
bis = ByteBuffer.wrap(bytes);
}
else {
bis = ByteBuffer.wrap(packet.getData());
}
fireProcessMessage(new UdpAddress(packet.getAddress(),
packet.getPort()), bis);
}
catch (SocketTimeoutException stex) {
// ignore
}
catch (PortUnreachableException purex) {
synchronized (DefaultUdpTransportMapping.this) {
listener = null;
}
logger.error(purex);
if (logger.isDebugEnabled()) {
purex.printStackTrace();
}
if (SNMP4JSettings.isFowardRuntimeExceptions()) {
throw new RuntimeException(purex);
}
break;
}
catch (SocketException soex) {
logger.error("Socket for transport mapping "+toString()+
" error: "+soex.getMessage(), soex);
stop = true;
}
catch (IOException iox) {
logger.warn(iox);
if (logger.isDebugEnabled()) {
iox.printStackTrace();
}
if (SNMP4JSettings.isFowardRuntimeExceptions()) {
throw new RuntimeException(iox);
}
}
}
synchronized (DefaultUdpTransportMapping.this) {
listener = null;
stop = true;
if (socket != null) {
socket.close();
}
}
------------------------------
Message: 2
Date: Tue, 03 Feb 2009 01:31:59 +0100
From: Frank Fock <[email protected]>
Subject: Re: [SNMP4J] DefaultUdpTransportMapping class and
ListenThread
To: Vivi Zhang <[email protected]>
Cc: [email protected]
Message-ID: <[email protected]>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi,
Since Socket.receive blocks until a packet is received,
I do not see any advantage of the suggested code.
Best regards,
Frank
Vivi Zhang wrote:
Hi,
I am looking at ways to tuning my application with embedded SNMP agent
with SNMP4J and SNMP4J-Agent.
I have two questions about the implementation of inner class
ListernThread.
Q1) Line 315 * try {
socket.receive(packet);
}
catch (InterruptedIOException iiox) {
if (iiox.bytesTransferred <= 0) {
continue;
}*
If we see SocketTimeoutException since there is no packet to read, can
we put the thread to a small break before it goes for another
socket.receive() call? The suggestion is intended to ease some
pressure on the CPU and network resource. Do you see any potential
drawback from this change?
Q2) Line 307 *DatagramPacket packet = new DatagramPacket(buf, buf.length,
udpAddress.getInetAddress(),
udpAddress.getPort());**
*A DatagramPacket object is created repeatedly within the while
loop. What is the drawback if we move the stmt outiside while loop. The
logic behind this suggestion is avoiding object creating repeatedly
withing the loop.
Thanks.
Vivi
P.S.
I am using SNMP4J 1.9.3c.
Following is a section of code I cut from DefaultUdpTransportMapping.java.
============================================================================
while (!stop) {
* DatagramPacket packet = new DatagramPacket(buf, buf.length,
udpAddress.getInetAddress(),
udpAddress.getPort());*
try {
try {
socket.receive(packet);
}
catch (InterruptedIOException iiox) {
if (iiox.bytesTransferred <= 0) {
* continue;*
}
}
if (logger.isDebugEnabled()) {
logger.debug("Received message from "+packet.getAddress()+"/"+
packet.getPort()+
" with length "+packet.getLength()+": "+
new OctetString(packet.getData(), 0,
packet.getLength()).toHexString());
}
ByteBuffer bis;
// If messages are processed asynchronously (i.e. multi-threaded)
// then we have to copy the buffer's content here!
if (isAsyncMsgProcessingSupported()) {
byte[] bytes = new byte[packet.getLength()];
System.arraycopy(packet.getData(), 0, bytes, 0, bytes.length);
bis = ByteBuffer.wrap(bytes);
}
else {
bis = ByteBuffer.wrap(packet.getData());
}
fireProcessMessage(new UdpAddress(packet.getAddress(),
packet.getPort()), bis);
}
catch (SocketTimeoutException stex) {
// ignore
}
catch (PortUnreachableException purex) {
synchronized (DefaultUdpTransportMapping.this) {
listener = null;
}
logger.error(purex);
if (logger.isDebugEnabled()) {
purex.printStackTrace();
}
if (SNMP4JSettings.isFowardRuntimeExceptions()) {
throw new RuntimeException(purex);
}
break;
}
catch (SocketException soex) {
logger.error("Socket for transport mapping "+toString()+
" error: "+soex.getMessage(), soex);
stop = true;
}
catch (IOException iox) {
logger.warn(iox);
if (logger.isDebugEnabled()) {
iox.printStackTrace();
}
if (SNMP4JSettings.isFowardRuntimeExceptions()) {
throw new RuntimeException(iox);
}
}
}
synchronized (DefaultUdpTransportMapping.this) {
listener = null;
stop = true;
if (socket != null) {
socket.close();
}
}
_______________________________________________
SNMP4J mailing list
[email protected]
http://lists.agentpp.org/mailman/listinfo/snmp4j
_______________________________________________
SNMP4J mailing list
[email protected]
http://lists.agentpp.org/mailman/listinfo/snmp4j
