Tuxedo Client using async calls... hangs sometimes!

Hello, I've noticed a fancy behaviour on Tuxedo Client applications...
I wrote an application which receives multiple connections on a socket and dispatches messages to a server with asynchronous calls, then receives asynchronous notifications. Of course it can send multiple async requests before getting a notification.
It seems that when I send an async request, and then I send another async request without getting the reply to the previous one first, the system hangs...
The call stack is quite clear... and I can see on the unix message queue an outstanding message..
I've used the TPU_DIP method, because none of the other unsolicited flavours works:
TPU_SIG: Hangs because malloc can be interrupted by a signal, and within the unsol handrel a tpalloc (which in turn calls malloc) is called, and a kernel spinlock is locked
TPU_THREAD: Hangs exactly in the same way ad TPU_DIP, but on a different thread...
Does anyone knows how to solve this HUGE issue? Thanks in advance!
Marco Romagnuolo
0 0x3000403abc8 in _tmmsgrcv(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so#1 0x3000405e650 in _tmrcvunsol(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so
#2 0x3000405d2f8 in tpchkunsol(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so
#3 0x3000405e480 in _tmchk4unsol(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so
#4 0x30004036ee4 in _tmatmienter(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so
#5 0x30004022944 in tpallocinternal(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so
#6 0x300040226f0 in tpalloc(...) in /usr/tuxedo/bea/tuxedo8.1/lib/libtux.so

I need asynchronous notifications for their particular behaviour: they are notified as soon as they arrive through signals ot through a thread handled by Tuxedo itself.
Now I've changed the policy of the process to use tpgetrply, but I have to poll in order to make it work, and it is UGLY.... anyway... anything else doesn't seem to work

Similar Messages

Using Async calls in a Util class

I have a Utility class that I want to put code in that I'm reusing over and over again. This includes Async.handleEvent calls. If I call an instance of the Util class from a [Test(async)] method, can I use the Async call in that other event?
When I tried this in my code, it said that it "Cannot add asynchronous functionality to methods defined by Test,Before or After that are not marked async" but my [Test] is marked as async. And when I was in the same class, I was able to have non-Test methods make Async calls.
Thanks for the help!
Mocked up Example (please note these are two separate files):
// Test.as
package
    import TimerUtil;
    public class Test
        [Test(async)]
        public function waitForTimer():void
            TimerUtil.newTimer();
// TimerUtil.as
package
    import flash.events.TimerEvent;
    import flash.utils.Timer;
    import org.flexunit.async.Async;
    public class TimerUtil
        private static var instance:TimerUtil;
        private var timer:Timer;
        private static const TIME_OUT:int = 1000;
        public static function newTimer() : void
            if (!instance) instance = new TimerUtil();
            instance.timer = new Timer(1000, 5);
            instance.timer.start();
            instance.waitForFifthTick();
        private function waitForFifthTick() : void
            Async.handleEvent(this, instance.timer, TimerEvent.TIMER, handleTimerTick, TIME_OUT);
        private function handleTimerTick ( evt:TimerEvent, eventObject:Object) : void
            if ( timer.currentCount == 5 )
                // We are at the Fifth click, move on.
            else
                Async.handleEvent(this, instance.timer, TimerEvent.TIMER, handleTimerTick, TIME_OUT);
Message was edited by: zyellowman2
I missed a comma. And a few "instance"s

Yes, you can put async calls in utility classes or wherever you want. That isn't the issue here though.
In your test you call TimerUtil.newTimer()
That code creates a timer and then returns. Then your test returns. At that moment, FlexUnit sees that your test has finished and that there are no outstanding registered asynchronous events so it declares the test a success. Then, a second later, your other method executes and tries to register an async call. FlexUnit has no idea what test this pertains to, or why this call is being made as it is no longer in the context of the original test, hence the message you are receiving that you can only using Async calls inside of a method marked with a Async... as, in FlexUnit's opion, the method that is executing this code no longer has anything to do with a test with an async param.
The best way to think about this is a chain that cannot be broken. As soon as a method finishes executing with, FlexUnit will declare it a success unless we are waiting on something else (an async event of some sort) That is what the Async syntax really does, it tells the framework not to declare this method a success just yet, and that there is something else we need to account for to make that decision.
So, right now, in your newTimer() you don't let the framework know that you will be waiting for an async event. Therefore, when that method and the test method is over, so is the test.
Mike

Client (Solaris) remote call hangs 225sec if Server (Window7) is down.

We are noticing the RMI Client remote call from Solaris machine hang of exactly 225 seconds if RMI Server on Windows7 machine is shutdown. RMI Server is shutdown for valid reasons. The expectation is, Client would throw a RemoteException immediately for futher processing. The Client hang for 225 minutes is causing performance issues. This has been consistently reproduced across multiple machines with Client on Solaris and Server on Windows 7. There are no 225 sec wait issues if server is on Windows XP or Solaris.
Usecase (java version "1.6.0_14")
1. Start RMI Server on Windows 7 machine
2. Start RMI Client on Solaris machine
3. Wait till client connects to the server and makes some remote call every 1sec.
4. Shutdown RMI Server.
5. Client waits 225 seconds before a time out error is thrown.
Has anyone seen this problem? Any solutions to get around this?
Code used to test this
RmiTestServer.java
public class RmiTestServer
public static void main(String args[])
try
java.rmi.registry.Registry zRegistry = null;
try
zRegistry = java.rmi.registry.LocateRegistry.createRegistry ( 1099 );
catch ( Exception zEx )
zRegistry = java.rmi.registry.LocateRegistry.getRegistry ( 1099 );
if ( zRegistry != null )
System.out.println( "Binding RmiTestService..." );
RmiTest c = new RmiTestImpl();
zRegistry.rebind ( "RmiTestService", c );
System.out.println( "Done Binding RmiTestService..." );
else
System.out.println( "Registry is null." );
catch (Exception e)
e.printStackTrace();
RmiTest.java
public interface RmiTest extends java.rmi.Remote
public void callHello() throws java.rmi.RemoteException;
RmiTestImpl.java
public class RmiTestImpl extends java.rmi.server.UnicastRemoteObject
implements RmiTest
public RmiTestImpl() throws java.rmi.RemoteException
super();
public void callHello() throws java.rmi.RemoteException
System.out.println ("Hello World!!!");
RmiTestClient
public class RmiTestClient
public static void main(String[] args)
System.out.println ("Connecting to: " + args[0] + " Port 1099" );
long lTime = System.currentTimeMillis();
try
RmiTest c = null;
java.rmi.registry.Registry registry = java.rmi.registry.LocateRegistry.getRegistry ( args[0], 1099 );
if ( registry!=null )
System.out.println ("Start lookup... " );
c = (RmiTest)registry.lookup ( "RmiTestService" );
System.out.println( "Lookup Done - " + (System.currentTimeMillis() - lTime)/1000 + "sec\n" );
while( true )
lTime = System.currentTimeMillis();
System.out.println( "Calling Server..." );
c.callHello();
System.out.println( "Done Calling Server - " + (System.currentTimeMillis() - lTime)/1000 + "sec\n" );
lTime = System.currentTimeMillis();
Thread.sleep(1000);
catch (Exception e)
System.out.println( "Time for Exception - " + (System.currentTimeMillis() - lTime)/1000 + "sec\n" );
e.printStackTrace();
Output
Calling Server...
Jun 9, 2011 10:59:19 PM sun.rmi.transport.tcp.TCPChannel newConnection
FINE: main: reuse connection
Jun 9, 2011 10:59:19 PM sun.rmi.transport.tcp.TCPChannel free
FINE: main: reuse connection
Done Calling Server - 0sec
Calling Server...
Jun 9, 2011 10:59:20 PM sun.rmi.transport.tcp.TCPChannel newConnection
FINE: main: reuse connection
Jun 9, 2011 10:59:20 PM sun.rmi.transport.tcp.TCPChannel free
FINE: main: reuse connection
Done Calling Server - 0sec
Calling Server...
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPConnection isDead --------------->RMI Server Shutdown
FINER: main: exception:
java.net.SocketException: Connection reset
at java.net.SocketInputStream.read(SocketInputStream.java:168)
at java.io.BufferedInputStream.fill(BufferedInputStream.java:218)
at java.io.BufferedInputStream.read(BufferedInputStream.java:237)
at sun.rmi.transport.tcp.TCPConnection.isDead(TCPConnection.java:174)
at sun.rmi.transport.tcp.TCPChannel.newConnection(TCPChannel.java:173)
at sun.rmi.server.UnicastRef.invoke(UnicastRef.java:110)
at RmiTestImpl_Stub.callHello(Unknown Source)
at RmiTestClient.main(RmiTestClient.java:23)
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPConnection isDead
FINE: main: server ping failed
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPChannel free
FINE: main: close connection
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPConnection close
FINE: main: close connection
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPChannel createConnection
FINE: main: create connection
Jun 9, 2011 10:59:21 PM sun.rmi.transport.tcp.TCPEndpoint newSocket
FINER: main: opening socket to [<IP Address>:65427]
Jun 9, 2011 10:59:21 PM sun.rmi.transport.proxy.RMIMasterSocketFactory createSocket
FINE: main: host: <IP Address>, port: 65427
Jun 9, 2011 10:59:24 PM sun.rmi.transport.tcp.TCPChannel$1 run
FINER: RMI Scheduler(0): wake up
Jun 9, 2011 10:59:24 PM sun.rmi.transport.tcp.TCPChannel$1 run
FINER: RMI Scheduler(0): wake up
Jun 9, 2011 10:59:39 PM sun.rmi.transport.tcp.TCPChannel$1 run
FINER: RMI Scheduler(0): wake up
Jun 9, 2011 10:59:39 PM sun.rmi.transport.tcp.TCPChannel freeCachedConnections
FINER: RMI Scheduler(0): connection timeout expired
Jun 9, 2011 10:59:39 PM sun.rmi.transport.tcp.TCPConnection close
FINE: RMI Scheduler(0): close connection
Time for Exception - 224sec
java.rmi.ConnectException: Connection refused to host: <IP Address>; nested exception is:
java.net.ConnectException: Connection timed out
at sun.rmi.transport.tcp.TCPEndpoint.newSocket(TCPEndpoint.java:601)
at sun.rmi.transport.tcp.TCPChannel.createConnection(TCPChannel.java:198)
at sun.rmi.transport.tcp.TCPChannel.newConnection(TCPChannel.java:184)
at sun.rmi.server.UnicastRef.invoke(UnicastRef.java:110)
at RmiTestImpl_Stub.callHello(Unknown Source)
at RmiTestClient.main(RmiTestClient.java:23)
Caused by: java.net.ConnectException: Connection timed out
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.PlainSocketImpl.doConnect(PlainSocketImpl.java:333)
at java.net.PlainSocketImpl.connectToAddress(PlainSocketImpl.java:195)
at java.net.PlainSocketImpl.connect(PlainSocketImpl.java:182)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:366)
at java.net.Socket.connect(Socket.java:529)
at java.net.Socket.connect(Socket.java:478)
at java.net.Socket.<init>(Socket.java:375)
at java.net.Socket.<init>(Socket.java:189)
at sun.rmi.transport.proxy.RMIDirectSocketFactory.createSocket(RMIDirectSocketFactory.java:22)
at sun.rmi.transport.proxy.RMIMasterSocketFactory.createSocket(RMIMasterSocketFactory.java:128)
at sun.rmi.transport.tcp.TCPEndpoint.newSocket(TCPEndpoint.java:595)
... 5 more

Could you check to see if the remote database is up and running before calling the remote database procedure?

Tuxedo client calls Tuxedo server using FML32

I used a Tuxedo client calling a Tuxedo Server.
Tuxedo server:
It receives FML32 buffer and parse the info from ui_fml.h, e.g.
#define     UI_SID     ((FLDID32)33852433)     /* number: 298001     type: long */
#define     UI_NAME     ((FLDID32)33852434)     /* number: 298002     type: long */
#define     UI_TEST     ((FLDID32)33852435)     /* number: 298003     type: long */
In server code:
void TEST(TPSVCINFO *rqst) {
char line[200];
long len = sizeof(line);
FBFR32 bfr = (FBFR32) rqst->data;
Fget32(bfr, UI_TEXT, 0, line, (FLDLEN32*)&len);
printf("Fstrerror32(Ferror32)=[%s]\n", Fstrerror32(Ferror32);
In Tuxedo client code:
int main(int argc, char* argv[]) (
char *sendbuf;
if (tpinit((TPINIT *) NULL) == -1) {
          (void) fprintf(stderr, "tpcall2X:Tpinit failed\n");
          exit(1);
if((sendbuf = (FBFR32 *) tpalloc("FML32", NULL, BUFLEN)) == (FBFR32*) NULL) {
          (void) fprintf(stderr,"Error allocating send buffer\n");
          tpterm();
          exit(1);
len = Fsizeof32(sendbuf);
if (-1 == Finit32(sendbuf, (FLDLEN32)len)) {
          tpfree((char*)sendbuf);
          exit(1);
ret = tpcall("TEST", (char *)sendbuf, 0L, &rcvbuf, &rcvlen, 0);
The server returns "Fstrerror32(Ferror32)=[LIBFML_CAT:2: ERROR: Buffer not fielded]", while the client works well.
What are the reason? How to change that?
Thanks.

Hi Bill,
Your problems are mostly in the client. You allocate sendbuf as an FML32 buffer, but then proceed to strcpy() into it which won't work. To manipulate the contents of an FML/FML32 buffer you must use the FML routines. So removing the strcpy(sendbuf, temp) in the client and replacing it with Fadd32(sendbuf, UI_TEXT, temp, sendlen) allows your application to work.
By the way, in general you shouldn't need to be doing all the memset() calls you are doing. The extraneous stuff in buffers will be ignored, and certainly never memset() an FML32 buffer as it is not simply a bunch of bytes, but a complex dynamic structure that Tuxedo maintains.
Finally on the server side, I couldn't get this line to compile as I indicated earlier:
memset(out.out, 0, 4096);
so I don't know how that is compiling for you, but the GCC compiler will not accept that, nor does it even make sense.
Regards,
Todd Little
Oracle Tuxedo Chief Architect
PS Here is the exact code that worked on my system:
Server:
#include <string.h>
#include <stdio.h>
#include <ctype.h>
#include <atmi.h>     /* TUXEDO Header File */
#include <userlog.h>     /* TUXEDO Header File */
#include "bill.h"
#if defined(__STDC__) || defined(__cplusplus)
tpsvrinit(int argc, char *argv[])
#else
tpsvrinit(argc, argv)
int argc;
char **argv;
#endif
     /* Some compilers warn if argc and argv aren't used. */
     argc = argc;
     argv = argv;
     /* userlog writes to the central TUXEDO message log */
     userlog("Welcome to the simple server");
     return(0);
#ifdef __cplusplus
extern "C"
#endif
void
#if defined(__STDC__) || defined(__cplusplus)
TEST(TPSVCINFO *rqst)
#else
TEST(rqst)
TPSVCINFO *rqst;
#endif
char line[200];
char *out;
long n,len;
FBFR32 *bfr = (FBFR32*) rqst->data;
out = tpalloc("X_OCTET", NULL, 4096);
memset(out, 0, 4096);
memset(line , 0, sizeof(line));
Fget32(bfr, UI_TEXT, 0, line, (FLDLEN32*)&len);
printf("Fstrerror32(Ferror32)=[%s]\n", Fstrerror32(Ferror32));
printf("line=[%s]\n", line);
n = sprintf(out, "success");
tpreturn(TPSUCCESS, 0, out, n, 0);
}And the client:
#include <string.h>
#include <stdio.h>
#include <atmi.h>
#include "bill.h"
#define BUFLEN 2400
int main(int argc, char* argv[])
char *sendbuf, *rcvbuf, service[15];
long sendlen, rcvlen;
int ret, i;
char *result;
char temp[2000];
FLDLEN32 len = 0;
if (tpinit((TPINIT *) NULL) == -1) {
(void) fprintf(stderr, "tpcall2X:Tpinit failed\n");
exit(1);
if((sendbuf = (FBFR32 *) tpalloc("FML32", NULL, BUFLEN)) == (FBFR32*) NULL) {
(void) fprintf(stderr,"Error allocating send buffer\n");
tpterm();
exit(1);
if((rcvbuf = (char *) tpalloc("X_OCTET", NULL, BUFLEN)) == NULL) {
(void) fprintf(stderr,"Error allocating receive buffer\n");
tpfree(sendbuf);
tpterm();
exit(1);
len = Fsizeof32(sendbuf);
if (-1 == Finit32(sendbuf, (FLDLEN32)len))
tpfree((char*)sendbuf);
tpfree((char*)rcvbuf);
exit(1);
/* memset(rcvbuf, 0, sizeof(rcvbuf));
memset(service, 0, sizeof(service));
memset(temp, 0, sizeof(temp));
strcpy(service, "TEST");
strcat(temp, "(FLDID(168070265)) test \n\n");
sendlen = (long)strlen(temp);
strcpy(sendbuf, temp);
Fadd32(sendbuf, UI_TEXT, temp, sendlen);
printf("TEST call [%s] begin.\n", service);
printf("sendbuf len=[%d]\n",sendlen);
ret = tpcall(service, (char *)sendbuf, 0L, &rcvbuf, &rcvlen, 0);
printf("ret=[%d]\n", ret);
if(ret == -1) {
printf("tpcall [%s] failed.\n", service);
printf("Tperrno = %d, %s\n", tperrno, tpstrerror(tperrno));
tpfree(sendbuf);
tpfree(rcvbuf);
tpterm();
exit(1);
printf("Returned string is: [%s]\n", rcvbuf);
printf("tpcall successs.\n");
/* Free Buffers & Detach from System/T */
tpfree(sendbuf);
tpfree(rcvbuf);
tpterm();
return(0);
}

HT1222 I have iphone 4 and the software is ios 7.0.2. I am facing problems with my phone. Since the last software update my phone has started giving problems. like it hangs sometime when i recieve an incoming call. The carrier status bar shifts down a bit

I have iphone 4 and the software is ios 7.0.2. I am facing problems with my phone. Since the last software update my phone has started giving problems. like it hangs sometime when i recieve an incoming call. The carrier status bar shifts down a bit and sometimes it vanishes automatically. I,m confused.
Secondly i have an update IOS 7.0.4 waiting in my Software Update menu. I have tried so many times to update my iphone but it always fails to update the IOS 7.0.4. Any body suggest anything regarding this

Wow. Lots of help. Thanks apple

Lync Client uses wrong RTP Ports for calls from/to RGS with Agent Anonymity

We have QoS implemented and client ports for audio, video und application defined by Set-CsConferencingConfiguration. We also use firewalls in our LAN between the different VLANs for Clients, Servers and Gateways/SBC. Only RTP from the client with the defined
ports are allowed by the firewall. Media ByPass is enabled.
In all normal cases, the right ports will be used and marked by GPO with the right DSCP value. But if an agent get a call from a RGS which has agent anonymity enabled, the client uses a port in the range 1024-65535 for audio. Also if you make a call on behalf
of the RGS, the client use a random port between 1024-65535. As soon, as the source of the call is in another VLAN (e.g. a call from PSTN which comes in over a SBC in e separate VLAN), the firewall between the two VLANs block the RTP traffic.
We see the deny on the firewall log and in the SBC log we see, the reinvite for the media by pass with the IP of the agent and a not valid port. We also see, that no RTP from the client/agent will arrive the SBC and no RTP from the SBC will arrive the client/agent.
So the call will be disconnected, as soon as an agent wants accept the call.
Is there an additional setting to make sure, the Lync client always use the valid RTP port range?
This behavior exist in Lync 2010 and Lync 2013 clients.

Hi Holger,
Thanks for reply!
Sure! I set all AudioPorts on all Services, but the problem are not the ports used by the server, the problem are the ports used by the client. We set the client ports to 49152 with a count of 40. The client (2013 and also 2010) use these ports correctly in all
cases exept for call from/on behalf of an RGS with Agent Anonymity.
If we disable the RGS agent anonymity, restart the client of the agent, then the client uses also the correct source ports for RTP.
I've checked this behaviour now on 3 customer installations, our own productive installation and in our lab.
Because until now only one of our customers have firewalls between the internal VLANs, only this single customer have the issues...
Regards,
Stephan

Using tuxedo client 8.1 with tuxedo server 6.4

Hi,
I am using Weblogic server 7 sp2 with Tuxedo 8.1 on my Windows XP machine, I have configured weblogic (config.xml) to use WTC to connect to tuxedo 6.4 on windows 2000 machine which in turn connects to tuxedo 6.4 server. When I try to do a tpcall I get an error on weblogic server log 'TPENOENT(6):0:0:TPED_MINVAL(0):QMNONE(0):0:Could not find service'.
My question is I haven't made any configuration changes on the tuxedo server end like ubb.config or dm.config files. We are currently in the process of upgrading to weblogic 7 or 8. At the moment we are using JOLT to connect to the existing tux services on the dynix box.
I want to know if I can configur WTC on weblogic 7 using tuxedo 8.1 API to connect to tuxedo 6.4. What all changes do I need to make on both WLS and Tuxedo end.
I have read some documentation which state ways where old tuxedo client versions 7.1 and earlier can access tuxedo server 8 by configuring WSH (CLOPT -t) option in ubb.config file.
Can someone please help.
Thanks,
Smita

Hello Smita,
I don't think WTC is supported with Tuxedo 6.4. I think (for a number of reasons) that you really should consider an upgrade to 6.5 or newer version of Tuxedo.
Regards,
/Per
Per Lindström - R2Meton AB, SWEDEN

Running Oracle Reports via a client using system.exec call from war file

Hi all,
I am trying to deploy a war file application to a J2EE 10.1.3 application server. The deployment successfully completes. However testing the application fails.
When testing, the application reveals that calling a Oracle Reports 6i client (using a cmd line exec) from the war file returns application errors. The errors we get from executing the call below (using debug statements), creates this error:
OperatingSystemCmd: cmd = D:\Oracle\Dev6i\BIN\rwcli60 MODULE="CLTMLST" USERID="<userid>/<pwd>@<db>" DESTYPE="FILE" DESFORMAT="PDF" DESNAME="<path>\<pdf filename>.PDF" PARAMFORM="NO" BACKGROUND="NO" SERVER="r6i.world" TOLERANCE=0 ADVCODE="<param>"
OperatingSystemCmd: Command returned 3
WrsRunReport: Done executing report
WrsRunReport: Error running report Return code 3
Error on Screen is REP-0178 : Cannot connect to reports service.
However running the same command from a operating system command prompt in windows reveals that the command runs successfully.
Is there any particular OC4J J2EE settings that are required to get this statement to run properly?
Any help is most appreciative.
Cheers
Rodney

Hi All,
From much research and experimenting to get this to work I looked at the java.lang.Runtime class and noticed that seperate processes can be started using this very important Java class.
The first thing that I experimented with was in regards to seeing the environment OC4J runs against. Using the java.lang.Runtime class I executed a standard "cmd /c set" command in my Windows environment, and noticed that Oracle Application Server uses its own environment and not a standard windows login environment for its OC4J containers. Apache Tomcat on the other hand uses the standard user login environment.
So to get the application to be able to work properly we needed to override the particular environment the process needed to be able to get it to work. There is a exec command which allows you to override completely the environment for a process you would like to run. This method call does not in any way shape or form change the standard OC4J environment. This was done by executing the same "set" but with the overriden environment. Note that no environment variables from the OC4J container are carried into this new environment.
Cheers
Rodney

Async call using ICallBack

hi, In an Async call, there is a ICallBack interface, which is having a init(ODIInformationObj),
handleCallBack(long pElapsedTime)
handleFaultCallback(long pElapsedTime,Exception pException).
can u pls help me how to invoke the handleCallBack() method, or else
it will be invoked internally through the Thread call.
pls suggest me to post the Message via(DirectConnection) and get back the operation information to an ODIService.
Note:
I have this code:
Message xmlMessage = XMLMessageFactory.getInstance().createMessage(payload1);
payload1="<ns1:ODIInvokerServiceResponse " + " xmlns:ns1=\"http://xmlns.oracle.com/ODIService\">\n" +
"<ns1:odiPlan> " + pOdiPlanName + "</ns1:odiPlan>\n" +
"<ns1:elapsedTime> " + pElapsedTime +
" </ns1:elapsedTime>\n" +
" </ns1:ODIInvokerServiceResponse>";
getting Direct connection using:
DirectConnection directconnection =
loc.createDirectConnection(compositeDN,
"MasterBPELProcessDirect");
Sending Message like:
directConnection.post(pOperation, pNm);

You Might Try tapping the Home button and see what that does, Do it on a Test call with one of your Relative or Friend's so they know that when you tap the button and if it happens to end the call they won't be concerned because it was a test Call..
Hope that helps b33

Disconnecting Tuxedo Clients

I would like to know if there is a simple way to disconnect Tuxedo clients from
a Tuxedo server (perhaps in tmadmin??). My preference is to be able to disconnect
a tuxedo client based on the user that is logged in (eg, the username listed when
a 'pclt' command is run in tmadmin).
I'm using version 7.1 of Tuxedo.
Any help appreciated.
Rod.

You can do this at runtime. You can either write a program that calls the MIB, or
use a ud32 script. You'll need to know the client ID and some other key fields.
If you get stuck, post another message and we'll see if someone can help you a
little more.
Rod Hamilton wrote:
Thanks Scott,
Is this something that can be done at runtime? Can I use a utility to set the
clients state to DEAD or do I need to write a custom app to easily do this?
Thanks,
Rod.
Scott Orshan <[email protected]> wrote:
You can use the MIB to set a Client's state to DEAD.
Rod Hamilton wrote:
I would like to know if there is a simple way to disconnect Tuxedoclients from
a Tuxedo server (perhaps in tmadmin??). My preference is to be ableto disconnect
a tuxedo client based on the user that is logged in (eg, the usernamelisted when
a 'pclt' command is run in tmadmin).
I'm using version 7.1 of Tuxedo.
Any help appreciated.
Rod.

Tuxedo clients

Hello, i am trying to build clients for tuxedo in multiple Operating Systems. So for Linux + Windows i can use Workstation Clients or Jolt clients to do my job
and call remotely tuxedo services. My question is if i can to do something similar to mobile OS like Android or iOS. Is there any way to create
tuxedo clients on mobile OS without the use of a web browser??
Thank you very much
Kostas

Hi Mats,
I have already seen the SALT. The problem is that i don't want to use SOAP in my application.
I want to create my own clients not based on any kind of browser and only communicating to the tuxedo server
directly and not in an intermediate Web Server like Web Logic.
So far, based on what i see this can only be done using CORBA. Of course there is a downside of
extending the Android SDK and possibly also the MAC OS and iOS as well.
That is my thoughts for now. Any correction, suggestion will be extremely helpful.
Best regards,
Kostas

Debugging a tuxedo service using dbx

Hi,
Is is possible to debug a tuxedo service using dbx by enabing the tmboot switches
and -g option of the compiler. I am able to step into tpsvrinit during initialisation,
but not able to do when a client calls a service. I could not step into the service
by setting breakpoints. Please help.
rgds,
Dominic

The only thing to be aware of is that while you play around in the debugger, Tuxedo
is timing out your transaction.
So, if you want things to keep working while you debug, setup some nice long timouts.
...Lyall
"RC Bryan" <[email protected]> wrote:
>
I have not done this on Solaris in a while but basically, what you have
to do is
to build the process with -g and start it with tmboot as you would normally.
After the process is running, you can attach by typing:
dbx name pid
where the name is the name of the executable and pid is the process id
of the
server process (obtained either with ps -ef | grep name or with verbose
mode psr
in tmadmin). This will break into the running process. You can then
set your
break points in the service routines as required. When the service is
entered,
control will return to your debugger session and you can debug as you
normally
would.
Incidentally, I find the buildserver -k (keep) option to be useful when
debugging
servers. This allows you to debug through the startup code that is normally
deleted
as a part of the buildserver process.
Regards,
/RC Bryan
"Dominic" <[email protected]> wrote:
Hi,
Is is possible to debug a tuxedo service using dbx by enabing the tmboot
switches
and -g option of the compiler. I am able to step into tpsvrinit during
initialisation,
but not able to do when a client calls a service. I could not step into
the service
by setting breakpoints. Please help.
rgds,
Dominic

Tuxedo client running under Windows CE

Does anyone know if the Tuxedo client portion can under on a Windows
CE device and make service calls?
(Windows CE 3.0)

Hi Rumcajs,
The current version of the LabVIEW PDA module (7.1) does not support
calls to external code in the form of .NET assemblies. However, it DOES
support calls to C-style DLLs and the way to do it is shown in the
following example:
Calling External Code in LabVIEW PDA for Pocket PC - Battery Information Example
Please let me know if you have any further questions, thanks.
Have fun!
- Philip Courtois, Thinkbot Solutions

Oracle Database Web Service Client using UTL_DBWS :: ORA-29532 Error

Hi,
I have the Oracle Database 10.2.0.1.0 :-
SQL> select * from v$version;
BANNER
Oracle Database 10g Enterprise Edition Release 10.2.0.1.0 - Prod
PL/SQL Release 10.2.0.1.0 - Production
CORE 10.2.0.1.0 Production
TNS for 32-bit Windows: Version 10.2.0.1.0 - Production
NLSRTL Version 10.2.0.1.0 - ProductionI have written a simple Web Services Client using the classes gfrom the UTL_DBWS package. I loaded the JAR file dbwsclient.jar in the SYS Schema and I am trying to use it in the USF Schema.
However, I have hit this error & I ma unable to proceed :-
SQL> select get_stock_price from dual;
select get_stock_price from dual
ERROR at line 1:
ORA-29532: Java call terminated by uncaught Java exception:
java.lang.IllegalAccessException: javax.xml.rpc.ServiceException:
java.security.AccessControlException: the Permission
(java.lang.RuntimePermission getClassLoader) has not been granted to USF. The
PL/SQL to grant this is dbms_java.grant_permission( 'USF',
'SYS:java.lang.RuntimePermission', 'getClassLoader', '' )
ORA-06512: at "USF.UTL_DBWS", line 193
ORA-06512: at "USF.UTL_DBWS", line 190
ORA-06512: at "USF.GET_STOCK_PRICE", line 17Can you please help me with this ?
Regards,
Sandeep

Hi,
The error message said
the Permission(java.lang.RuntimePermission getClassLoader) has not been granted to USF.
I'd follow the suggestion
The PL/SQL to grant this is dbms_java.grant_permission( 'USF','SYS:java.lang.RuntimePermission', 'getClassLoader', '' )
In case you have not done so, consult the Callout Users Guide @
http://www.oracle.com/technology/sample_code/tech/java/jsp/callout_users_guide.htm
Kuassi http://db360.blogspot.com

Error "The memory could not be "read" in tuxedo client 8.0

Old Environment:
1.Tuxedo Server 6.4
2.Tuxedo /WS 6.4
5.PowerBuilder 5.1
New Environment:
1.Tuxedo Server 8.0
2.Tuxedo Client 8.0
3.PowerBuilder 8
In my upgrade software version progress, I want to unchange my old source code
or little changed. I've DLL develop from Visual C++ to communicate bewteen Tuxedo
and PowerBuilder.
After upgraded software, I found memory error from my client PowerBuilder application
(See error from attach file.) and I don't know what's cause of this error. And
after I've found this error I try to use tuxedo /ws version 6.4 again, It's workable
and not found memory error.
Please give me if you found the solution of this case .
Thanks you very much for yours help.
[error.bmp]

We are working on a similar case from another customer. Please contact customer support and
provide them details of your case.
-Deepak
Sutep wrote:
Old Environment:
1.Tuxedo Server 6.4
2.Tuxedo /WS 6.4
5.PowerBuilder 5.1
New Environment:
1.Tuxedo Server 8.0
2.Tuxedo Client 8.0
3.PowerBuilder 8
In my upgrade software version progress, I want to unchange my old source code
or little changed. I've DLL develop from Visual C++ to communicate bewteen Tuxedo
and PowerBuilder.
After upgraded software, I found memory error from my client PowerBuilder application
(See error from attach file.) and I don't know what's cause of this error. And
after I've found this error I try to use tuxedo /ws version 6.4 again, It's workable
and not found memory error.
Please give me if you found the solution of this case .
Thanks you very much for yours help.
Name: error.bmp
error.bmp Type: Bitmap Image (image/bmp)
Encoding: base64

Tuxedo Client using async calls... hangs sometimes!

Similar Messages

Maybe you are looking for