Re: [Axis2] Memory leaks with 1.4.1
On Wed, Apr 8, 2009 at 11:34 AM, Amila Suriarachchi amilasuriarach...@gmail.com wrote: hi, you may get a snapshot build from the 1.5 branch here[1]. we are hoping to release an RC soon. thanks, Amila. What about backporting all these fixes to the 1.4 branch? I think that's the stable branch... and as I see, 1.5 seems to required JDK 1.5
Re: Axis2 MTOM and WSSecurity (encryption)
On Wed, Mar 25, 2009 at 6:13 PM, Erwin Reinhoud erwin.reinh...@ictu.nl wrote: Hello, I am wondering if more users need to use mtom and ws-security. There is an issue (3407) since dec 2007, but is still not fixed. Isn;t it one of the benefits that mtom supports ws-security? Kind regards, Erwin It seems that Sandesha and Rampart don't get too much attention...
Sandesha sequence auto termination
Hello, I found the following property: org.apache.sandesha2.storage.beans.RMSBean.avoidAutoTermination but I don't understand how it works AFAIS, the setter for this property is only called on the client side, so the client can ask Sandesha not to terminate the sequence? The problem seems to be that on the server side, the value of the property isn't set anywhere... so it remains false and the Sandesha server keeps sending TerminateSequence messages back to the client, as a response to the client's own TerminateSequence message(which probably is another bug)
Sandesha sequence auto termination
Hello, I found the following property: org.apache.sandesha2.storage.beans.RMSBean.avoidAutoTermination but I don't understand how it works AFAIS, the setter for this property is only called on the client side, so the client can ask Sandesha not to terminate the sequence? The problem seems to be that on the server side, the value of the property isn't set anywhere... so it remains false and the Sandesha server keeps sending TerminateSequence messages back to the client, as a response to the client's own TerminateSequence message(which probably is another bug)
Sandesha sequence auto termination
Hello, I found the following property: org.apache.sandesha2.storage.beans.RMSBean.avoidAutoTermination but I don't understand how it works AFAIS, the setter for this property is only called on the client side, so the client can ask Sandesha not to terminate the sequence? The problem seems to be that on the server side, the value of the property isn't set anywhere... so it remains false and the Sandesha server keeps sending TerminateSequence messages back to the client, as a response to the client's own TerminateSequence message(which probably is another bug)
Sandesha sequence auto termination
Hello, I found the following property: org.apache.sandesha2.storage.beans.RMSBean.avoidAutoTermination but I don't understand how it works AFAIS, the setter for this property is only called on the client side, so the client can ask Sandesha not to terminate the sequence? The problem seems to be that on the server side, the value of the property isn't set anywhere... so it remains false and the Sandesha server keeps sending TerminateSequence messages back to the client, as a response to the client's own TerminateSequence message(which probably is another bug)
Re: sandesha problem
Any thoughts? On Mon, Feb 2, 2009 at 11:13 AM, Andrei Ivanov andrei.iva...@gmail.com wrote: On Mon, Feb 2, 2009 at 8:07 AM, Amila Suriarachchi amilasuriarach...@gmail.com wrote: Can you try with the axis2 snapshot and sandesha2 snapshot. I see that sandesha2 snapshot is using generics... and I use java 1.4 Try using different INTERNAL_KEY s rather than the SEQUENCE_KEY How can I do that? AFAIS, in both 1.3 and in snapshot the internal key is defined like this: Sandesha2Constants.INTERNAL_SEQUENCE_PREFIX + : + to + : + sequenceKey I am providing a different sequence key for each sequence... so this should make the internal key different already. thanks, Amila. On Sun, Feb 1, 2009 at 9:17 PM, Andrei Ivanov andrei.iva...@gmail.com wrote: Hello, I'm trying to add WS-RM to my services and clients, but I'm having some problems. I followed the user guide and when I tried to do some testing, I noticed some exceptions being thrown. I've generated a client stub(PersonalDataServiceStub) from the services WSDL and I've added these lines to the end of the constructor(PersonalDataServiceStub(ConfigurationContext configurationContext, String targetEndpoint, boolean useSeparateListener)): final String sequenceKey = SandeshaUtil.getUUID(); clientOptions.setProperty(SandeshaClientConstants.SEQUENCE_KEY, sequenceKey); clientOptions.setProperty(SandeshaClientConstants.LAST_MESSAGE, Constants.VALUE_TRUE); Then, in the test class I created 2 instances of the stub and called some methods: ConfigurationContext configContext = ConfigurationContextFactory.createConfigurationContextFromFileSystem(CLIENT_REPO_PATH, axis2_xml); PersonalDataServiceStub stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); // param was initialized stub.getPersonalData(param); stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); stub.getFamilyTree(request); I'm setting the sequenceKey myself because I want to use 2 separate sequences for the calls, but something goes wrong. I used tcpmon to see what it was doing: Client: 1. CreateSequence 2. invoke getPersonalData 3. CreateSequence (for the 2nd call) 4. TerminateSequence (for the 1st call) 5. invoke getFamilyTree 6. TerminateSequence (for the 2nd call) Server: 1. CreateSequenceResponse 2. result for getPersonalData 3. CreateSequenceResponse 4. TerminateSequence (why is the server trying to end the 1st sequence itself?) 5. result for getFamilyTree 6. Fault, the action http://schemas.xmlsoap.org/ws/2005/02/rm/TerminateSequence; cannot be processed at the receiver (probably for the 1st TerminateSequence from the client) 7. SequenceFault, the 1st sequence is already terminated... probably at step 4 the server terminated it 8. TerminateSequence, for the 2nd sequence Am I doing something wrong? or maybe it is a bug in sandesha? remember that I want to use a different sequence for each operation. If you need additional data, please tell me. Thanks -- Amila Suriarachchi WSO2 Inc. blog: http://amilachinthaka.blogspot.com/
Re: sandesha problem
Anyone? On Mon, Feb 2, 2009 at 11:13 AM, Andrei Ivanov andrei.iva...@gmail.com wrote: On Mon, Feb 2, 2009 at 8:07 AM, Amila Suriarachchi amilasuriarach...@gmail.com wrote: Can you try with the axis2 snapshot and sandesha2 snapshot. I see that sandesha2 snapshot is using generics... and I use java 1.4 Try using different INTERNAL_KEY s rather than the SEQUENCE_KEY How can I do that? AFAIS, in both 1.3 and in snapshot the internal key is defined like this: Sandesha2Constants.INTERNAL_SEQUENCE_PREFIX + : + to + : + sequenceKey I am providing a different sequence key for each sequence... so this should make the internal key different already. thanks, Amila. On Sun, Feb 1, 2009 at 9:17 PM, Andrei Ivanov andrei.iva...@gmail.com wrote: Hello, I'm trying to add WS-RM to my services and clients, but I'm having some problems. I followed the user guide and when I tried to do some testing, I noticed some exceptions being thrown. I've generated a client stub(PersonalDataServiceStub) from the services WSDL and I've added these lines to the end of the constructor(PersonalDataServiceStub(ConfigurationContext configurationContext, String targetEndpoint, boolean useSeparateListener)): final String sequenceKey = SandeshaUtil.getUUID(); clientOptions.setProperty(SandeshaClientConstants.SEQUENCE_KEY, sequenceKey); clientOptions.setProperty(SandeshaClientConstants.LAST_MESSAGE, Constants.VALUE_TRUE); Then, in the test class I created 2 instances of the stub and called some methods: ConfigurationContext configContext = ConfigurationContextFactory.createConfigurationContextFromFileSystem(CLIENT_REPO_PATH, axis2_xml); PersonalDataServiceStub stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); // param was initialized stub.getPersonalData(param); stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); stub.getFamilyTree(request); I'm setting the sequenceKey myself because I want to use 2 separate sequences for the calls, but something goes wrong. I used tcpmon to see what it was doing: Client: 1. CreateSequence 2. invoke getPersonalData 3. CreateSequence (for the 2nd call) 4. TerminateSequence (for the 1st call) 5. invoke getFamilyTree 6. TerminateSequence (for the 2nd call) Server: 1. CreateSequenceResponse 2. result for getPersonalData 3. CreateSequenceResponse 4. TerminateSequence (why is the server trying to end the 1st sequence itself?) 5. result for getFamilyTree 6. Fault, the action http://schemas.xmlsoap.org/ws/2005/02/rm/TerminateSequence; cannot be processed at the receiver (probably for the 1st TerminateSequence from the client) 7. SequenceFault, the 1st sequence is already terminated... probably at step 4 the server terminated it 8. TerminateSequence, for the 2nd sequence Am I doing something wrong? or maybe it is a bug in sandesha? remember that I want to use a different sequence for each operation. If you need additional data, please tell me. Thanks -- Amila Suriarachchi WSO2 Inc. blog: http://amilachinthaka.blogspot.com/
Re: axis2 1.4.1 and rampart 1.4 issue
Don't top post. On Wed, Feb 4, 2009 at 10:51 AM, Iyengar, Kumar kumar_iyen...@bmc.com wrote: What missing jars? I added Rampart 1.4 mar file to the modules directory as per the instructions in the ReadMe file. I get the error: In your first mail, there are some exceptions like these: java.lang.ClassNotFoundException: org.apache.rampart.Rampart These tell you what classes are missing. If you search through your rampart jars, you'll find that rampart-core contains this class. Just copy it to the lib directory.
Axiom bug?
Hello,
I'm trying to use Sandesha with Rampart and I think I've found a bug.
When the server is replying to a CreateSequence message, it gets to
RampartSender,
which calls org.apache.rampart.MessageBuilder.build(MessageContext msgCtx),
which verifies that the security header is empty and tries to remove it:
if ( secHeader != null secHeader.isEmpty(doc) ) {
secHeader.removeSecurityHeader(doc);
}
WSSecHeader.removeSecurityHeader(Document doc) does this:
...
Node parent = securityHeader.getParentNode();
parent.removeChild(securityHeader);
Parent:
soapenv:Header
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
xmlns:wsa=http://www.w3.org/2005/08/addressing;
wsse:Security
xmlns:wsse=http://docs.oasis-open.org/wss/2004/01/oasis-200401-wss-wssecurity-secext-1.0.xsd;
soapenv:mustUnderstand=1 /
wsa:Tohttp://192.168.1.5:6060/axis2/services/personaldataserv...@286914895//wsa:To
wsa:ReplyTowsa:Addresshttp://www.w3.org/2005/08/addressing/none/wsa:Address/wsa:ReplyTo
wsa:MessageIDurn:uuid:E3FC5985317BCF16351233758322948/wsa:MessageID
wsa:Actionhttp://schemas.xmlsoap.org/ws/2005/02/rm/CreateSequenceResponse/wsa:Action
wsa:RelatesTourn:uuid:A07FCCC1972C7E06671233758319824/wsa:RelatesTo
/soapenv:Header
securityHeader:
wsse:Security
xmlns:wsse=http://docs.oasis-open.org/wss/2004/01/oasis-200401-wss-wssecurity-secext-1.0.xsd;
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
soapenv:mustUnderstand=1 /
Then, org.apache.axiom.om.impl.dom.ParentNode.removeChild(Node) is
called and it gets to this, begining with line 482:
if (this.firstChild == tempNode) {
// If this is the first child
this.firstChild = null;
this.lastChild = null;
tempNode.parentNode = null;
}
tempNode is the security header and it is equal to the firstChild of
the envelope header.
The firstChild is removed... then the lastChild(which is the relatesTo
child) is removed...
The resulting envelope header is this:
soapenv:Header
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
xmlns:wsa=http://www.w3.org/2005/08/addressing; /
I'm using Axis-1.4.1, Addressing-1.41, Sandesha2-1.3, Rampart-1.4, Axiom 1.2.8
I've looked in the Axiom sources and it seems this code is from before
revision 389047 (Moving doom into ws-commons axiom from axis2)
I think this is a bug...
Am I right? Should I create an issue?
Re: Axiom bug?
On Wed, Feb 4, 2009 at 5:41 PM, Andreas Veithen
andreas.veit...@gmail.com wrote:
Andrei,
What is the Axiom version you are using? If it is not 1.2.8 or a
recent snapshot, please check against version 1.2.8. If you can
confirm that the issue still exists in that version, please open a
JIRA issue in the WSCOMMONS project.
1.2.8... it say right at the end of my email...
I'll open an issue...
Thanks
Regards,
Andreas
On Wed, Feb 4, 2009 at 16:08, Andrei Ivanov andrei.iva...@gmail.com wrote:
Hello,
I'm trying to use Sandesha with Rampart and I think I've found a bug.
When the server is replying to a CreateSequence message, it gets to
RampartSender,
which calls org.apache.rampart.MessageBuilder.build(MessageContext msgCtx),
which verifies that the security header is empty and tries to remove it:
if ( secHeader != null secHeader.isEmpty(doc) ) {
secHeader.removeSecurityHeader(doc);
}
WSSecHeader.removeSecurityHeader(Document doc) does this:
...
Node parent = securityHeader.getParentNode();
parent.removeChild(securityHeader);
Parent:
soapenv:Header
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
xmlns:wsa=http://www.w3.org/2005/08/addressing;
wsse:Security
xmlns:wsse=http://docs.oasis-open.org/wss/2004/01/oasis-200401-wss-wssecurity-secext-1.0.xsd;
soapenv:mustUnderstand=1 /
wsa:Tohttp://192.168.1.5:6060/axis2/services/personaldataserv...@286914895//wsa:To
wsa:ReplyTowsa:Addresshttp://www.w3.org/2005/08/addressing/none/wsa:Address/wsa:ReplyTo
wsa:MessageIDurn:uuid:E3FC5985317BCF16351233758322948/wsa:MessageID
wsa:Actionhttp://schemas.xmlsoap.org/ws/2005/02/rm/CreateSequenceResponse/wsa:Action
wsa:RelatesTourn:uuid:A07FCCC1972C7E06671233758319824/wsa:RelatesTo
/soapenv:Header
securityHeader:
wsse:Security
xmlns:wsse=http://docs.oasis-open.org/wss/2004/01/oasis-200401-wss-wssecurity-secext-1.0.xsd;
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
soapenv:mustUnderstand=1 /
Then, org.apache.axiom.om.impl.dom.ParentNode.removeChild(Node) is
called and it gets to this, begining with line 482:
if (this.firstChild == tempNode) {
// If this is the first child
this.firstChild = null;
this.lastChild = null;
tempNode.parentNode = null;
}
tempNode is the security header and it is equal to the firstChild of
the envelope header.
The firstChild is removed... then the lastChild(which is the relatesTo
child) is removed...
The resulting envelope header is this:
soapenv:Header
xmlns:soapenv=http://schemas.xmlsoap.org/soap/envelope/;
xmlns:wsa=http://www.w3.org/2005/08/addressing; /
I'm using Axis-1.4.1, Addressing-1.41, Sandesha2-1.3, Rampart-1.4, Axiom
1.2.8
I've looked in the Axiom sources and it seems this code is from before
revision 389047 (Moving doom into ws-commons axiom from axis2)
I think this is a bug...
Am I right? Should I create an issue?
Re: sandesha problem
On Mon, Feb 2, 2009 at 8:07 AM, Amila Suriarachchi amilasuriarach...@gmail.com wrote: Can you try with the axis2 snapshot and sandesha2 snapshot. Try using different INTERNAL_KEY s rather than the SEQUENCE_KEY I've just tried the snapshots and I don't think there's any difference. The server side keeps sending TerminateSequence messages, as if it is ignoring the TerminateSequence from the client. Or maybe the server side sees the CreateSequence messages from the next sequence and somehow it decides it needs to terminate the first one itself? thanks, Amila. On Sun, Feb 1, 2009 at 9:17 PM, Andrei Ivanov andrei.iva...@gmail.com wrote: Hello, I'm trying to add WS-RM to my services and clients, but I'm having some problems. I followed the user guide and when I tried to do some testing, I noticed some exceptions being thrown. I've generated a client stub(PersonalDataServiceStub) from the services WSDL and I've added these lines to the end of the constructor(PersonalDataServiceStub(ConfigurationContext configurationContext, String targetEndpoint, boolean useSeparateListener)): final String sequenceKey = SandeshaUtil.getUUID(); clientOptions.setProperty(SandeshaClientConstants.SEQUENCE_KEY, sequenceKey); clientOptions.setProperty(SandeshaClientConstants.LAST_MESSAGE, Constants.VALUE_TRUE); Then, in the test class I created 2 instances of the stub and called some methods: ConfigurationContext configContext = ConfigurationContextFactory.createConfigurationContextFromFileSystem(CLIENT_REPO_PATH, axis2_xml); PersonalDataServiceStub stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); // param was initialized stub.getPersonalData(param); stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); stub.getFamilyTree(request); I'm setting the sequenceKey myself because I want to use 2 separate sequences for the calls, but something goes wrong. I used tcpmon to see what it was doing: Client: 1. CreateSequence 2. invoke getPersonalData 3. CreateSequence (for the 2nd call) 4. TerminateSequence (for the 1st call) 5. invoke getFamilyTree 6. TerminateSequence (for the 2nd call) Server: 1. CreateSequenceResponse 2. result for getPersonalData 3. CreateSequenceResponse 4. TerminateSequence (why is the server trying to end the 1st sequence itself?) 5. result for getFamilyTree 6. Fault, the action http://schemas.xmlsoap.org/ws/2005/02/rm/TerminateSequence; cannot be processed at the receiver (probably for the 1st TerminateSequence from the client) 7. SequenceFault, the 1st sequence is already terminated... probably at step 4 the server terminated it 8. TerminateSequence, for the 2nd sequence Am I doing something wrong? or maybe it is a bug in sandesha? remember that I want to use a different sequence for each operation. If you need additional data, please tell me. Thanks -- Amila Suriarachchi WSO2 Inc. blog: http://amilachinthaka.blogspot.com/
sandesha problem
Hello, I'm trying to add WS-RM to my services and clients, but I'm having some problems. I followed the user guide and when I tried to do some testing, I noticed some exceptions being thrown. I've generated a client stub(PersonalDataServiceStub) from the services WSDL and I've added these lines to the end of the constructor(PersonalDataServiceStub(ConfigurationContext configurationContext, String targetEndpoint, boolean useSeparateListener)): final String sequenceKey = SandeshaUtil.getUUID(); clientOptions.setProperty(SandeshaClientConstants.SEQUENCE_KEY, sequenceKey); clientOptions.setProperty(SandeshaClientConstants.LAST_MESSAGE, Constants.VALUE_TRUE); Then, in the test class I created 2 instances of the stub and called some methods: ConfigurationContext configContext = ConfigurationContextFactory.createConfigurationContextFromFileSystem(CLIENT_REPO_PATH, axis2_xml); PersonalDataServiceStub stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); // param was initialized stub.getPersonalData(param); stub = new PersonalDataServiceStub(configurationContext, targetEndpoint, true); stub.getFamilyTree(request); I'm setting the sequenceKey myself because I want to use 2 separate sequences for the calls, but something goes wrong. I used tcpmon to see what it was doing: Client: 1. CreateSequence 2. invoke getPersonalData 3. CreateSequence (for the 2nd call) 4. TerminateSequence (for the 1st call) 5. invoke getFamilyTree 6. TerminateSequence (for the 2nd call) Server: 1. CreateSequenceResponse 2. result for getPersonalData 3. CreateSequenceResponse 4. TerminateSequence (why is the server trying to end the 1st sequence itself?) 5. result for getFamilyTree 6. Fault, the action http://schemas.xmlsoap.org/ws/2005/02/rm/TerminateSequence; cannot be processed at the receiver (probably for the 1st TerminateSequence from the client) 7. SequenceFault, the 1st sequence is already terminated... probably at step 4 the server terminated it 8. TerminateSequence, for the 2nd sequence Am I doing something wrong? or maybe it is a bug in sandesha? remember that I want to use a different sequence for each operation. If you need additional data, please tell me. Thanks
