-----Original Message-----
From: Brian Ko [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 11, 2003 12:20 PM
To: '[EMAIL PROTECTED]'
Subject: RE: Slow instantiation of the Service class in clientI am a newbie in Axis. But I am also working on message style service.I tested your program (without service installed) in window 2000 environment but the response was much quicker than 8 seconds. That may mean that the slowness is from your service, not the client. If you can share your service class code and wsdd file, I am willing to test it here.Brian Ko
-----Original Message-----
From: Trieu, Jason T - CNF [mailto:[EMAIL PROTECTED]
Sent: Tuesday, March 11, 2003 12:02 PM
To: '[EMAIL PROTECTED]'
Subject: Slow instantiation of the Service class in clientI am developing a test message-style AXIS client & server and notice that the client takes about 8 seconds for the client to instantiate the Service class on an IBM RS6000 AIX4.3 machine. Is this quite normal or am I doing something wrong? I used the provided sample code for message-style clients to create mine. Below is the code. Appreciate any insight in this.
---------------------------------------------------------------------------------------------------------------------------
package ping.client;
import javax.xml.parsers.*;
import org.w3c.dom.*;
import org.apache.axis.client.*;
import org.apache.axis.message.*;
import org.apache.axis.utils.*;
import java.io.*;
import java.text.*;
import java.net.URL;
import java.util.*;public class pingClient {
public void sendIt(String args[]) throws Exception {
String endpointURL = "http://"+args[0]+":8080/axis/services/pingService";
// Create Call object (via Service) to call server
// ****** This statement takes 8 seconds to complete *************
Service sv = new Service();
Call call = (Call) sv.createCall();
call.setTargetEndpointAddress(new URL(endpointURL));
// Create a SOAP body elements to send to server
SOAPBodyElement[] Selems = new SOAPBodyElement[2];// Build DOM Document via Factory and builder
DocumentBuilderFactory fact = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = fact.newDocumentBuilder();// Now create DOM doc to use for creating DOM elements
Document DOMdoc = builder.newDocument();// Create SOAP element using DOM doc
Element ping = DOMdoc.createElementNS("cnf:dms","PING");
org.w3c.dom.Text pingdata = DOMdoc.createTextNode("are you there?");
ping.appendChild(pingdata);
Selems[0] = new SOAPBodyElement(ping);// Create SOAP element using XMLUtils
Selems[1] = new SOAPBodyElement(XMLUtils.StringToElement("cnf:dms",
"PONG","let me know"));
// Call the sever
Vector resp = (Vector) call.invoke(Selems);// Get SOAP elemements and DOM elements
SOAPBodyElement retSOAPel = null;
Element retDOMel = null;for (int i = 0; i<resp.size();i++) {
retSOAPel = (SOAPBodyElement) resp.get(i);
retDOMel = retSOAPel.getAsDOM();
System.out.println("Element "+i+": "+XMLUtils.ElementToString(retDOMel));
}
}
public static void main(String[] args) throws Exception {
if (args.length < 1) {
System.out.println("Bad Host");
System.exit(1);
}(new pingClient()).sendIt(args);
}
} // End of class
Title: Slow instantiation of the Service class in client
Hi
Guys,
I had
the same problem in Windows 2000 runing on a 500MHz P3 or a 1.1GHz K6AMD. I saw
it happen in either a Sun rpc service client or an Axis one. The call object
instantiation was very quick. The best I could do was pool the service
object so I did not have to worry about serialization in the architecture
slowing things down. I am using an Applet and people told me that using that
combined with the "Big" download of Axis jars to support the client API of Axis
was the problem. I solved my problem by going to a much lighter and very much
faster implementation of client-side SOAP. My service time is now "0sec" on the
same hardware and I actually can smile again.
I
would be interested if you find out the issue. Perhaps Brian has a stand-alone
application and Jason a mobile code type application?
Best
Regards,
-Tony
- Slow instantiation of the Service class in client Trieu, Jason T - CNF
- RE: Slow instantiation of the Service class in... Brian Ko
- RE: Slow instantiation of the Service class in... Anecito, Anthony (HQP)
- RE: Slow instantiation of the Service class in... Trieu, Jason T - CNF
- RE: Slow instantiation of the Service class in... Anecito, Anthony (HQP)
- RE: Slow instantiation of the Service class in... Brian Ko
