Fix
Pack Notes for
WebSphere Business Integration Adapter Framework Version 2.6.0.12
The patch notes contain the following sections:
This patch contains enhancements and fixes for the following customer-reported and internally discovered issues.
Adapter Framework runtime changes
|
Problem tracking number |
As of Version |
APAR Number |
Problem |
|
91673 |
2.6.0.12 |
JR28596 |
Symptom:
When stopping a connector, you might get many AMQ9209 errors in the MQ error log. |
|
Resolution: The Adapter Framework code has been changed so that some MQ connections will be closed before connectors stop. |
|||
|
2.6.0.12 |
JR28595 |
Symptom:
If IDL is the transport, when the attribute value starts with '\0', Adapter
Framework converts it into CxIgnore, which might not be desirable. |
|
|
Resolution: The ReplaceCxIgnore property lets you choose whether to
convert an attribute that starts with '\0' into CxIgnore. |
|||
|
90457 |
2.6.0.12 |
JR27465 |
Symptom: The business object
element parameter 'size' is not a mandatory attribute. If the 'size'
parameter is not set, Adapter Framework constructs a child business object
with single cardinality, even though it should be a multi-cardinality child business object. Adapter
Framework also does not throw any errors. |
|
Resolution: Adapter Framework code has been changed so that the 'size' attribute in a business object schema is a mandatory attribute. The following error is generated if the 'size' attribute is not set: 84035 |
|||
|
96336 |
2.6.0.12 |
JR29225 |
Symptom:
Adapters can not be shut down gracefully if InterChange Server is running with
RBAC enabled. |
|
Resolution:
Adapter Framework code has been changed so that adapters can be shut down
gracefully even when InterChange Server is running with RBAC enabled. |
|||
|
75843 |
2.6.0.12 |
JR30375 |
Symptom: The adapter can not be stopped as a Windows®
service when it is automatically restarted after a retry restart. The related trace
messages are as below: [Thread: CxSCManagerThread (#***)] [Type: Trace]
[Mesg: :setAppDomainState] |
|
Resolution:
When an adapter is registered as a Windows service, some variables used by
CxSCManagerThread are not correctly set after the adapter is auto-restarted
and this leads to an adapter termination failure. Adapter Framework code has
been changed so that the adapter can be stopped successfully in this case. |
|||
|
90866 |
2.6.0.12 |
- |
Symptom:
If the adapter receives an exception when calling the data handler DataHandler.createDataHandler(),
it does not print any messages. |
|
Resolution:
Adapter Framework code has been changed so that if the adapter receives an
exception when calling the data handler DataHandler.createDataHandler(),
the adapter log can print a message with a message ID from 9087 to 9090. |
|||
|
92084 |
2.6.0.12 |
- |
Adapter
runtime currency support for Windows Server 2008. |
|
92087 |
2.6.0.12 |
- |
Adapter
runtime currency support for HP_UX 11i V3 for PA-RISK. |
|
92097 |
2.6.0.12 |
- |
Adapter
runtime has updated its WBIA API to 1.8.1. |
|
92103 |
2.6.0.12 |
- |
Adapter
runtime has updated embedded IBM JRE/JDK to V1.4.2SR11. |
|
92104 |
2.6.0.12 |
- |
Adapter
runtime has certificated with IBM JRE/JDK V5.0SR7. |
|
92105 |
2.6.0.12 |
- |
Adapter
runtime has updated its embedded CEI to V6.0.0.13. |
|
69706 |
2.6.0.11 |
JR26745 |
Symptom: When InterChange Server is the broker, if an adapter on AIX® or Linux® starts without –c, the adapter log file is overwritten each time the adapter is restarted. |
|
Resolution: The Adapter Framework code has been changed so that the adapter log file content is not overwritten each time the adapter is restarted. |
|||
|
89409 |
2.6.0.11 |
JR27019 |
Symptom: When InterChange Server is the broker and JMS is the transport, if the adapter starts immediately after it is stopped by connector_manager_<connector> -stop/-stopgraceful, the adapter takes too much time (up to several minutes) to start again. The related traces are as follows: [Type: Trace] [Mesg: Maintain can not complete, waiting.....] |
|
Resolution: Both Adapter Framework code and InterChange Server code have been changed so that when Adapter Framework shuts down, it notifies InterChange Server. Therefore, InterChange Server does not need to wait two minutes until a timeout and can immediately accept the maintain message from Adapter Framework when Adapter Framework starts again. *The fix in InterChange Server code is in InterChange Server 4.3.0.4 LAIF10 and currently NOT included in InterChange Server 4.3.0.5. |
|||
|
89092 |
2.6.0.11 |
JR26923 |
Symptom: When InterChange Server is the broker and MQ is the transport, the runtime error that occurs in the adapter doVerbFor method is not reflected in the adapter log and trace and thus is hard to troubleshoot. |
|
Resolution: Adapter Framework code has been changed to catch the runtime error that occurs in doVerbFor method. Adapter Framework logs the error, and because it is a non-recoverable error, Adapter Framework exits after logging it. |
|||
|
90033 |
2.6.0.11 |
JR27299 |
Symptom: When InterChange Server is the broker and MQ is the transport, if the queue manager used between Adapter Framework and InterChange Server is not available during the adapter restarts, the adapter remains in pause state forever. The related logs are as follows: [Type: Fatal Error] [MsgID: 9032] [Mesg: Failed to obtain a connection with MQSeries, queue manager name = **, channel = **, host = **, port = **.] [Type: Info] [MsgID: 17064] [Mesg: Connector Agent state is paused.] |
|
Resolution: Adapter Framework code has been changed to log an error message and exit because failing to connect to the queue manager when the adapter restarts/starts up is a fatal error. If the reason for the MQ exception is not “Connection to queue manager lost”, Adapter Framework directly logs the message and exits. If the reason is “Connection to queue manager lost”, Adapter Framework tries 10 times to reconnect to the queue manager and if all attempts fail, it logs an error message and exits. |
|||
|
89968 |
2.6.0.11 |
JR27363 |
Symptom: When JMS is the transport, Adapter Framework uses sessions stored in the session pool to send a message to the replyToQueue after finishing the doVerbFor method. If for some reason the sessions in the session pool are not valid, Adapter Framework fails to send a message to the replyToQueue. And it creates a new session and succeeds to send a message to the fault queue. The related logs are as follows: [Type: Error] [MsgID: 9080] [Mesg: The replyToQueue is not valid or available, the message is sent to fault queue. Exception: javax.jms.JMSException: MQJMS2008: failed to open MQ queue **. Message ID: **] |
|
Resolution: Adapter Framework code has been changed so that after receiving a JMS exception when sending message to the replyToQueue, Adapter Framework creates a new session to send the message again. If it fails again, Adapter Framework logs the message and sends it to the fault queue. |
|||
|
2.6.0.11 |
(Enhancement) |
The customized logging is supported to record time related information on Adapter Framework synchronous calls. |
|
|
2.6.0.11 |
(Enhancement) |
Adapter Framework supports optionally writing AP queue messages as non-persistent. |
|
|
70108 |
2.6.0.11 |
- |
The CEI client embedded in Adapter Framework has been upgraded to CEI client version 5.1.0.3 |
|
70115 |
2.6.0.11 |
- |
The CwDataHandler.jar file in Adapter Framework has been upgraded to include the data handler fix for CR85920 |
|
88613 |
2.6.0.9 |
JR26451 |
Symptom: When brokerType is set to WAS in the adapter configuration file, Adapter Framework changes attribute 2147483647 of String type to CxBalnk when processing a request message. |
|
Resolution: Adapter Framework code has been changed so that it does not convert attribute 2147483647 of String type to CxBlank. It keeps it as 2147483647. |
|||
|
88429 |
2.6.0.9 |
JR26456 |
Symptom: When the adapter is started on the UNIX platform, it runs in background mode by default. If you modify the script to make it run in the front, ADK throws NullPointerException when the adapter restarts. The related error messages are as follows: [Thread: wbia_main (#***)] [Mesg: java.lang.NullPointerException at AppSide_Connector.AppEnd.run(AppEnd.java(Compiled Code))… |
|
Resolution: Adapter Framework code has been changed so that it does not throw NullPointerException when the adapter restarts in the manner described. |
|||
|
2.6.0.9 |
JR26574 |
Symptom: Adapter Framework start-up scripts do not allow JVM parameters to be set in adapter specific startup files on the UNIX platform. |
|
|
Resolution: Adapter Framework code has been changed so that you can set customized JVM parameters when starting the adapter on the UNIX platform. |
|||
|
88594 |
2.6.0.9 |
JR26434 |
Symptom: “connector_manager –stat” does not use appropriate exit codes. |
|
Resolution: Adapter Framework code has been changed. For the “connector_manager_<connector> -stat” command, Adapter Framework uses exit code ‘0’ when the adapter is running, and it uses exit code ‘1’ when the adapter is not running or is restarting. |
|||
|
69414 |
2.6.0.9 |
(Enhancement) |
An MQ/JMS linked exception is included in the trace file |
|
2.6.0.9 |
(Enhancement) |
Local transaction with WebSphere MQ JMS is enabled to avoid message duplication. |
|
|
69364 |
2.6.0.9 |
- |
Adapter runtime supports for SUSE 10 |
|
69369 |
2.6.0.9 |
- |
Adapter runtime supports for Red Hat Enterprise Linux 5 |
|
69087 |
2.6.0.8 |
JR25698 |
Symptom: For a stand-alone adapter on the UNIX platform, the .pid file is not created after it is started by the connector_manager script. Therefore, the stat command does not reflect the correct status of the adapter, nor can it reflect the correct result when the adapter has been shut down and the .pid file still exists. |
|
Resolution: Adapter Framework code has been changed so that the stat command reflects the correct status of the adapter. |
|||
|
2.6.0.8 |
JR26104 |
Symptom: The Clarify adapter hangs when it is shut down with system manager or by the service being stopped. |
|
|
Resolution: Adapter Framework code has been changed so that the Clarify adapter can be stopped successfully with system manager or by stopping the service. |
|||
|
88208 |
2.6.0.8 |
JR26150 |
Symptom: When JMS is the transport, if the adapter restarts whiling processing messages in the request queue, messages are left in an uncommitted state in the request queue. |
|
Resolution: Adapter Framework code has been changed so that messages are not left in an uncommitted state in the request queue. |
|||
|
87515 |
2.6.0.8 |
JR25833 |
Symptom: If the -c property is missing while starting a stand-alone adapter, a misleading message occurs: [Mesg: Default configuration file ** is missing, please copy your InterchangeSystem.cfg from your ICS installation.] |
|
Resolution: Adapter Framework has changed the message as follows [Mesg: Default configuration file ** is missing, if broker type is WMQI/WAS, please use -c parameter to specify the adapter configuration file; if broker is ICS, please copy your InterchangeSystem.cfg into the WebSphere Business Integration adapter installation directory.] |
|||
|
2.6.0.8 |
JR25819 |
Symptom: When InterChange Server is the broker and MQ is the transport, the adapter fails to deliver the message to the broker when using MQ with a remote queue. The related logs are as follows: [Type: Fatal Error] [MsgID: 17052] [Mesg: Failed to establish a session with MQ or to subscribe to queue AP/**/**.] |
|
|
Resolution: Adapter Framework code has been changed so that the adapter can deliver a message to the broker successfully when using MQ with a remote queue. |
|||
|
88080 |
2.6.0.8 |
JR26120 |
Symptom: When InterChange Server is the broker and IDL is the transport, the InterChange Server audit file is incomplete and unclear. It only states which user logged in or out. It does not state which adapter logged in or out. The related logs in the InterChange Server audit file are as follows: [Type: Info] [MsgID: 81100] [Mesg: User ** logged in.] |
|
Resolution: Both Adapter Framework code and InterChange Server code have been changed so that the InterChange Server audit file logs adapter infomation. The related logs in the InterChange Server audit file are as follows: [Type: Info] [MsgID: 81200] [Mesg: Adapter ** logged in as **.] Note: The fix in InterChange Server code is in InterChange Server version 4.3.0.5. |
|||
|
2.6.0.8 |
(Enhancement) |
First Failure Date Capture (FFDC) has been enabled in WebSphere Business Integration Adapter Framework. |
|
|
69163 |
2.6.0.8 |
- |
The version of JRE/SDK embedded in WebSphere Business Integration Adapter Framework has been upgraded to version 1.4.2SR7. |
|
87420 |
2.6.0.8 |
- |
WebSphere Business Integration Adapter Framework Tools provide limited support for Windows Vista |
|
87419 |
2.6.0.8 |
- |
Adapter Framework runtime currency support for Windows Vista is needed. |
|
85309 |
2.6.0.7 |
JR25307 |
Symptom: When JMS is the transport and InterChange Server is the broker, the adapter takes 20 minutes to shut down after it loses the connection with the database |
|
Resolution: Adapter Framework code has been changed so that the adapter takes less time (approximately 5 minutes) to shut down when the database connection is lost. |
|||
|
2.6.0.7 |
JR25311 |
Symptom: When JMS is the transport, ADK does not populate the required verb attribute on the child business object when putting the message on the DeliveryQueue |
|
|
Resolution: Adapter Framework code has been changed. A new parameter, -xPopChildVerb, has been added to set the default verb on the child business object if it is not set. See the “Additional discussion” section that follows for detail usage information. |
|||
|
2.6.0.7 |
JR25440 |
Symptom: When borkerType is set to WAS in the adapter configuration file, if the attribute value of String type is NULL, Adapter Framework converts it to “” when converting the business object to an XML message. |
|
|
Resolution: Adapter Framework code has been changed so that NULL is converted to CxIgnore. If you still want NULL to be converted to “”, you must use the “-xNullToBlank” parameter when starting the adapter. See the “Additional discussion” section that follows for detail usage information. |
|||
|
86981 |
2.6.0.7 |
JR25519 |
Symptom: The adapter fails to start when the Windows service is automatic. |
|
Resolution: Adapter Framework code has been changed so that the adapter starts successfully when the Windows service is automatic. |
|||
|
87208 |
2.6.0.7 |
JR25610 |
Symptom: When a NullPointerException is generated while parsing a message from request queue the resulting error message is incomplete. The related logs are as follows: [Thread: ***CONNECTOR/REQUESTQUEUEJmsThreadPool.thread* (#***)] [Type: Error] [MsgID: 0] [Mesg: null] |
|
Resolution: Adapter Framework code has been changed so that the error message is complete and indicates that a NullPointerException is generated when processing the element. The related logs are as follows: [Thread: ***CONNECTOR/REQUESTQUEUEJmsThreadPool.thread* (#***)] [Type: Error] [MsgID: 0] [Mesg: java.lang.NullPointerException while proceesing element: **] |
|||
|
69055 |
2.6.0.7 |
- |
Symptom: When JMS is the transport and the JMS message cannot be converted to a business object, the message is sent to the fault queue. But the fault message does not include the JMS customer properties. |
|
Resolution: Adapter Framework code has been changed so that the fault message includes the JMS customer properties. |
|||
|
86633 |
2.6.0.7 |
- |
ADK supports the CEI server that comes with WebSphere Process Server |
|
68997 |
2.6.0.7 |
- |
The JRE/SDK embedded in WebSphere Business Integration Adapter Framework has been upgraded to version 1.4.2SR6 |
|
2.6.0.7 |
(Enhancement) |
The adapter can be shut down gracefully if DeliveryTransport is set to JMS in the adapter configuration file. |
|
|
85776 |
2.6.0.6 |
JR24871 |
Symptom: Under special circumstances (when multiple CPU machines are used), two events may get the same ObjectEventID. |
|
Resolution: Adapter Framework code has been changed so that the ObjectEventID is unique for every event. |
|||
|
84100 |
2.6.0.6 |
JR23728 |
Symptom: When MQ is the transport, the adapter pauses for one minute while delivering a message to InterChange Server when the queue depth does not exceed 90 percent of the maximum depth. The related logs are as follows: [Mesg: Poll Thread is Sleeping for 1 minute as current MQ Depth exceeds 90 percentile of Maximum Depth] |
|
Resolution: Adapter Framework code has been changed so that the adapter does not pause when the queue depth does not exceed 90 percent of the maximum depth. |
|||
|
68129 |
2.6.0.6 |
JR23778 |
Symptom: When network problems between the adapter and an application occur, it takes 30 minutes to put the request messages into the fault queue. |
|
Resolution: Adapter Framework code has been changed so that it does not need so much time to put the request messages into the fault queue. |
|||
|
83747 |
2.6.0.6 |
JR23542 |
Symptom: When InterChange Server is the broker and the adapter is configured with PPD>1, multiple failed flows are generated when the adapter slave thread crashes while processing request messages. |
|
Resolution: Adapter Framework code has been changed so that no failed flows are generated when the slave thread crashes while processing request messages. |
|||
|
83647 |
2.6.0.6 |
JR23473 |
Symptom: When InterChange Server is the broker and the adapter is configured with PPD > 1 and AgentTraceLevel is set to 5, all the events in the event table are archived with the status “Skipped” and are not sent to the broker. |
|
Resolution: Adapter Framework code has been changed so that all the events are processed. |
|||
|
83678 |
2.6.0.6 |
JR23496 |
Symptom: When InterChange Server is the broker and MQ is the transport and the adapter is in restart/retry mode, for every retry, a new channel is opened but the channels opened in the previous retry are not closed. |
|
Resolution: Adapter Framework code has been changed so that the old channels can be used. |
|||
|
83778 |
2.6.0.6 |
JR23558 |
Symptom: When InterChange Server is the broker and the adapter on the UNIX platform is in retry mode, using “connector_manager –stop” gives a misleading message that indicates that the adapter has been stopped while it is still restarting. After restarting successfully, using “connector_manager –start” leads to multiple instances running. |
|
Resolution: Adapter Framework code has been changed so that using “connector_manager –stop” gives a correct message, such as “the adapter is waiting for restarting, so it can’t be stopped.” And using “connector_manager –start” later does not lead to multiple instances running. |
|||
|
68015 |
2.6.0.6 |
JR23481 |
Symptom: When JMS is the transport and the “reply to queue” of a request message is set to “put disable”, the replied message will be lost. It is neither sent to the fault queue nor rolled back to the request queue. |
|
Resolution: Adapter Framework code has been changed so that the replied message is sent to the fault queue. |
|||
|
68031 |
2.6.0.6 |
JR23500 |
Symptom: When JMS is the transport and the “reply to queue” of a request message is set to “put disable”, the replied message is sent to the fault queue. But the priority of the replied message in the fault queue has been changed to 4, which is not consistent with the value of the request message. |
|
Resolution: Adapter Framework code has been changed so that the priority of the replied message in the fault queue does not change. |
|||
|
68172 |
2.6.0.6 |
JR23901 |
Symptom: When InterChange Server is the broker and the adapter is configured to log files with limited file size and limited number of archives, the log file increases without archiving the files after starting the adapter without the “–c” parameter. |
|
Resolution: Adapter Framework code has been changed so that the log file can be archived with limited file size and limited number of archives. |
|||
|
83817 |
2.6.0.6 |
JR23579 |
Symptom: When InterChange Server is the broker and the adapter is configured to log files, the adapter logs to an incorrect path after starting the adapter without the “–c” parameter. |
|
Resolution: Adapter Framework code has been changed so that the adapter logs to the correct path. |
|||
|
84352 |
2.6.0.6 |
JR23880 |
Symptom: When JMS is the transport and the adapter property XMLNameSpaceFormat is set to “no”, Adapter Framework cannot process request messages without a namespace after starting adapter with “-xAFNamespaceAware=false.” The related logs are as follows: [MsgID: 0] [Mesg: String index out of range: -1] [Type: Error] [MsgID: 9068] [Mesg: The received message cannot be converted to business object Message. The message is not valid and is NOT processed. The response will be sent to the ReplyToQueue if it is set. Otherwise, the message will be sent to the fault queue….] |
|
Resolution: Adapter Framework code has been changed so that Adapter Framework processes request messages without a namespace. |
|||
|
84547 |
2.6.0.6 |
JR24080 |
Symptom: The adapter has two doVerbFor threads, and if both of them fail, Adapter Framework restarts the adapter in two threads, resulting in a NullPointerException. |
|
Resolution: Adapter Framework code has been changed so that the adapter can be restarted successfully without a NullPointerException and no messages in log file showing that multiple threads tried to restart the adapter. |
|||
|
84364 |
2.6.0.6 |
JR23896 |
Symptom: When JMS is the transport, if Adapter Framework picks up the Malformed data message that JMSType is null from the request queue, it ignores the message. The related logs are as follows: [Type: Error] [MsgID: 9085] [Mesg: REQUESTQUEUE can not be used to process Admin message. This Admin message is ignored. Message ID: ID:*] |
|
Resolution: Adapter Framework code has been changed so that the malformed data message that JMSType is null is put into the fault queue instead of being ignored as a misplaced administrative message in the request queue. |
|||
|
83918 |
2.6.0.6 |
JR23640 |
Symptom: When the agent and controller trace level is set to 0, the following message is written in the trace file when version changes are rolled back. If you have a high volume of transactions, it can cause a lot of traces and can also degrade performance. The related traces are as follows: [Type: Trace] [Mesg: :class com.ibm.adapters.datahandlers.xml.soap: DataHandler Version: Unknown] |
|
Resolution: Adapter Framework code has been changed so that the trace message is written only once when version changes are rolled back. |
|||
|
83732 |
2.6.0.6 |
JR23533 |
Symptom: When JMS is the transport, the adapter shuts down if the JMS provider shuts down. But the adapter is unable to reconnect and restart after the JMS provider restarts. The related traces are as follows: [Type: Trace] [Mesg: :error occurred when closing the queueconnectionjava.util.ConcurrentModificationException] |
|
Resolution: Adapter Framework code has been changed so that the adapter can reconnect and restart without a ConcurrentModificationException occurring. |
|||
|
83221 |
2.6.0.6 |
JR23439 |
Symptom: When starting an adapter by Cwservice, the adapter cannot bind to port when it is restarted. The related logs are as follows: [Type: Error] [MsgID: 9202] [Mesg: Failed to bind to port localhost: *] [Type: Error] [MsgID: 9203] [Mesg: Failed to open port *.] |
|
Resolution: Adapter Framework code has been changed so that the adapter is restarted successfully without errors occurring. |
|||
|
84473 |
2.6.0.6 |
JR24237 |
Symptom: When the IsRequired field in the business object is checked and the default value is set to CxBlank, CxBlank, not an empty string, is printed in the XML message. |
|
Resolution: Adapter Framework code has
been changed so that the empty string is printed in the XML message. |
|||
|
84957 |
2.6.0.6 |
JR24319 |
Symptom: When an adapter is started without a configuration file, the log file is rotated when the current log file is full, whereas the trace file continues to increase in size. |
|
Resolution: Adapter Framework code has been changed so that the trace file is rotated when the current trace file is full. |
|||
|
85243 |
2.6.0.6 |
JR24489 |
Symptom: When an adapter is started without a configuration file, it does not use the locale set in the configuration file from the InterChange Server repository, but instead uses the operating system locale. |
|
Resolution: Adapter Framework code has been changed so that the locale set in the configuration file from the InterChange Server repository is used. |
|||
|
85294 |
2.6.0.6 |
JR25090 |
Symptom: When the adapter receives a statusCode of 404 and returns an error string to Adapter Framework, the request message is locked in the request queue and the adapter terminates with an error like “Fault queue is either not valid or available.” |
|
Resolution: Adapter Framework code has been changed so that Adapter Framework adds the error string received from the adapter into the user header “description” of the message, and then puts the message into the fault queue. |
|||
|
2.6.0.6 |
JR24057 |
Symptom: When InterChange Server shuts down or the controller stops, the adapter can restart and retry to connect only five times before abandoning the operation. |
|
|
Resolution: Adapter Framework code has been changed by the addition of the –xMaxStartRetry parameter. -xMaxStartRetry=-1 (or any negative integer) The adapter retries to connect to the controller until it succeeds. -xMaxStartRetry=n (n>0) The adapter retries n times to connect to the controller. The interval between attempts is one minute. -xMaxStartRetry=0 The adapter does not retry to connect to the controller. Note: If the option is not set, the adapter retries to connect to the controller for a maximum of three times. |
|||
|
68724 |
2.6.0.6 |
JR24663 |
Symptom: When JMS is the transport, if the request message and the business object definition do not match, the request message is not put into the ReplyToQueue or fault queue. The related logs are as follows: [Type: Error] [MsgID: 9068] [Mesg: The received message cannot be converted to business object Message. The message is not valid and is NOT processed. The response will be sent to the ReplyToQueue if it is set. Otherwise, the message will be sent to the fault queue. Message ID: ID:*] [Type: Error] [MsgID: 9080] [Mesg: The replyToQueue is not valid or available, the message is sent to fault queue. Exception: javax.jms.InvalidDestinationException: MQJMS2008: failed to open MQ queue WBIMB.IDOC.RQST.STATUS.RESP. Message ID: ID:*] |
|
Resolution: Adapter Framework code has been changed so that the mismatched message is put into the fault queue if the ReplyToQueue is either invalid or unavailable. |
|||
|
68764 |
2.6.0.6 |
JR24751 |
Symptom: If PollFrequency is greater than RestartRetryInterval, the adapter fails to restart due to a NullPointerException. |
|
Resolution: Adapter Framework code has been changed so that the adapter restarts successfully |
|||
|
2.6.0.6 |
- |
Library packages xwbase.jar, xworacle.jar, xwsqlserver.jar and xwutil.jar have expired under the <fix_pack_install_location>/lib directory. They have been removed because these embedded JDBC drivers are not necessary for Adapter Framework and are not required for Adapter for JDBC. |
|
|
2.6.0.5 |
JR23106 |
Symptom: The master process works independently of the slave processes. Even when all slave processes have stopped, the master process continues to run and the connector agent shows up as “running” in the System Manager. |
|
|
Resolution: Adapter Framework code has been changed so that the master process starts a new slave process of the type that stopped. And if all slave processes stop and cannot restart successfully in the number of times specified in RestartRetryCount, the master process shuts down the adapter and System Manager is alerted. |
|||
|
82969 |
2.6.0.5 |
JR23053 |
Symptom: When InterChange Server is the broker and MQ is the transport, the adapter generates an authentication error when it starts. The related logs are as follows: [Type: Fatal Error] [MsgID: 9032] [Mesg: Failed to obtain a connection with MQSeries, queue manager name = *, channel = *, host =*, port = *. [info]::MQSeries:Completion code 2 Reason code 2035]. MQ error 2035 is an authentication error. |
|
Resolution: Adapter Framework code has been changed so that the adapter can start successfully. |
|||
|
82963 |
2.6.0.5 |
JR23042 |
Symptom: When InterChange Server is the broker and the adapter is running with PPD>1 and processing requests, all the slaves are stopped and some business objects are sent to the adapter. These business objects are returned as failed. During the restart of the slaves, the submitted business objects are also returned as failed. |
|
Resolution: Adapter Framework code has been changed so that it marks the adapter with a status of “Paused” to prevent the coming business objects from InterChange Server. |
|||
|
73853 |
2.6.0.5 |
JR23105 |
Symptom: When InterChange Server is the broker and the adapter is running with PPD>1, the following error occurs when the adapter starts up. The related logs are as follows: [Type: Fatal Error] [MsgID: 14001] [Mesg: Cannot start connector. Reason: [Type: Error] [MsgID: 17500] [Mesg: The parallel connector agent master was unable to spawn agent request slave processes. Reason: An error was encountered while trying to set up the resource cache for the connector agent. Reason: Checking to see if Recovery is needed...]] |
|
Resolution: Adapter Framework code has
been changed to fix four problems for this issue: |
|||
|
82985 |
2.6.0.5 |
JR23042 |
Symptom: When InterChange Server is the broker and the adapter is running with PPD>1, if one slave shuts down and then restarts, all the other slaves are still active but stop processing events until the stopped slave is fully restarted and active again. |
|
Resolution: Adapter Framework code has been changed so that the other slaves do not stop processing events when one shuts down and then restarts. |
|||
|
82923 |
2.6.0.5 |
JR23003 |
Symptom: A Java™ level deadlock occurs when PPD >1 and one slave process stops. |
|
Resolution: Adapter Framework code has been changed so that the deadlock problem is resolved. |
|||
|
82940 |
2.6.0.5 |
JR23940 |
Symptom: When the adapter restarts, Adapter Framework asks the adapter to close the connection with the application before shutting down. If the adapter fails to do so, Adapter Framework creates a new thread to do it. Adapter Framework generates a NullPointerException because there are two threads restarting the adapter at the same time. |
|
Resolution: Adapter Framework code has been changed so that Adapter Framework will try three times to close the connection with the application instead of creating a new thread. If all tries fail, Adapter Framework shuts down the adapter instead of restarting. Because the adapter may not start successfully when the connection with the application is not closed, the adapter must be restarted manually. The related logs are as follows: 17072 Failed while try to close connection with application, retry it after 60 seconds. 17073 Adapter can’t close connection with application, restart failed. Shutdown the adapter. |
|||
|
74545 |
2.6.0.5 |
JR23492 |
Symptom: When InterChange Server is the broker and JMS is the transport, the adapter generates a NullPointerException while connecting to MQ. |
|
Resolution: Adapter Framework code has been changed so that there is not a NullPointerException generated while connecting to MQ. |
|||
|
67922 |
2.6.0.5 |
JR23007 |
Symptom: When stopping the adapter using the adminmgr.sh file on AIX, the following error occurs: The JVM option is invalid: -server Could not create the java virtual machine. |
|
Resolution: Adapter Framework code has been changed to remove “-server” from the adminmgr.sh file. |
|||
|
55989 |
2.6.0.5 |
JR22894 |
Symptom: When starting an adapter that is already running, a new PID file overwrites the existing PID file. |
|
Resolution: Adapter Framework code has been changed so that when starting an adapter, if another adapter with the same name is running, the message “The Connector (connector_name) is already running” is printed instead of starting another process and producing a new PID file. |
|||
|
67835 |
2.6.0.5 |
- |
Symptom: The adapter fails to start due to an incorrect queue name written in the configuration file. But when it tries to restart after the time specified by the RestartRetryInterval within the RestartRetryCount, a Nullpointexception is generated. The related logs are as follows: [Type: Info] [MsgID: 17514] [Mesg: About to begin restart retry # 1 for serial connector agent *Connector.] [Type: Info] [MsgID: 17516] [Mesg: Waiting for configured restart retry interval: * minute(s) before restarting connector agent *Connector.] [System: Server] [Thread: Restart Thread - JMSConnector- retry #1 (#-2109935140)] [Mesg: java.lang.NullPointerException *] |
|
Resolution: Adapter Framework code has been changed so that if the adapter fails to start due to an incorrect queue name, it can restart normally after the time specified by the RestartRetryInterval within the RestartRetryCount. |
|||
|
74177 |
2.6.0.5 |
JR23358 |
Symptom: When the adapter is configured for event processing on delimited text and one of the text fields contains trailing escaped characters, the ArrayIndexOutOfBoundsException is generated. The related traces are as follows: [Type: Trace] [Mesg: :class com.crossworlds.DataHandlers.text.delimited :Error occured in getBO method: java.lang.ArrayIndexOutOfBoundsException] |
|
Resolution: Adapter Framework code has been changed so that when trailing escaped characters are used in the text fields, no ArrayIndexOutOfBoundsException occurs. |
|||
|
67066 |
2.6.0.5 |
JR21904 |
Symptom: When a string contains delimiter characters, the data handler cannot properly convert it to a business object. |
|
Resolution: Adapter Framework code has been changed so that a string containing delimiter characters can be converted properly to a business object. |
|||
|
83146 |
2.6.0.5 |
JR23146 |
Symptom: Unable to view business object structure through the Log Viewer when the log and trace file written in the configuration file points to the same file. |
|
Resolution: Adapter Framework code has been changed so that business object structure can be viewed through the Log Viewer. |
|||
|
72719 |
2.6.0.5 |
JR21558 |
Symptom: Adapter Framework does not support retry mechanism when using “adminmgr.sh –cConnector.cfg –s.” |
|
Resolution: Adapter Framework code has been changed so that it supports the retry mechanism. |
|||
|
74226 |
2.6.0.5 |
(Enhancement) |
If two ObjectEventIds are the same, the current business object represents a duplicate event. Adapter Framework will add warning message with dump BO in log. |
|
72183 |
2.6.0.4 |
JR22842 |
Symptom: When MQ is the transport, there is a ClassCastException when delivering a Business Object (BO) to InterChange Server. The related logs are as follows: [Type: Trace] [Mesg: :Posting BusObj to Interchange.] java.lang.ClassCastException: CxCommon.CxListBuffer |
|
Resolution: Adapter Framework code has been changed so that there is no ClassCastException. |
|||
|
61774 |
2.6.0.4 |
JR22843 |
Symptom: When InterChange Server is the broker and the adapter is started with –c, the controller and agent trace levels are not dynamically updated. A warning with MsgID14211 is logged in the adapter trace file that states “Dynamic update is not allowed.” |
|
Resolution: Adapter Framework code has been changed so that the controller and agent trace levels are dynamically updated |
|||
|
2.6.0.4 |
JR22327/JR22844 |
Symptom: When the request message generated by WebSphere MQ Integrator Broker (WMQI) has no namespace declaration, the message cannot be converted to a business object. The error description is similar to “The received message cannot be converted to business object. The message is not valid.” |
|
|
Resolution: A new parameter called AFNamespaceAware has been introduced. See the “Additional discussion” section for details about usage. |
|||
|
67476 |
2.6.0.4 |
JR22845 |
Symptom: The polling message is printed in the adapter trace file when the agent trace level is 1. The related traces are as follows: [Type: Trace] [Mesg: :[Type: Trace] [MsgID: 17008] [Mesg: Polling every 10000 milliseconds.] ] [Type: Trace] [Mesg: :[Type: Trace] [MsgID: 17009] [Mesg: Polling for subscription events.] ] |
|
Resolution: Adapter Framework code has been changed so that the polling message is not printed when the agent trace level is 1. |
|||
|
72403 |
2.6.0.4 |
JR22122 |
Symptom: The adapter fails to parse an incomplete XML message to a business object and when it logs a null message to InterChange Server, an IDL error occurs and the adapter state is set to “not connected.” Therefore, subsequent event delivery fails. The related traces are as follows: [Type: Trace] [Mesg: :IDL error encountered when processing logMsg: org.omg.CORBA.BAD_PARAM: Null Param (3) - wstring minor code: 4942F203 completed: Maybe] |
|
Resolution: Adapter Framework code has been changed so that the IDL error does not occur. |
|||
|
2.6.0.4 |
JR22846 |
Symptom: A local configuration file is needed for logging when InterChange Server is the broker |
|
|
Resolution: Both Adapter Framework code and InterChange Server code have been changed so that the local configuration file is not needed. To implement this fix, you must have InterChange Server version 4.3.0.2 or later version installed. |
|||
|
73521 |
2.6.0.4 |
JR22847 |
Symptom: The adapter shuts down with error 17069 if it cannot find its message file. The following misleading message is generated. [Type: Error] [MsgID: 17069] [Mesg: Connector Agent message file *** does not exist.] |
|
Resolution: Adapter Framework code has been changed so that MsgID 17069 is modified as follows. [Type: Error] [MsgID: 17069] [Mesg: Can’t find Message File *** - Aborting.] |
|||
|
71274 |
2.6.0.4 |
JR22849 |
Symptom: When the stand-alone adapter is started and stopped by the connector_manager script, the .pid file is not deleted. When the adapter runs “connector_manager –stat” later, it gives incorrect results. |
|
Resolution: Adapter Framework code has been changed so that the .pid file is deleted when the adapter stops, and “connector_manager –stat” can give correct results. |
|||
|
71268 |
2.6.0.4 |
JR21551 |
Symptom: When the stand-alone adapter processes MQDEAD message in the request queue, the invalid message stays in the request queue and the fault queue. |
|
Resolution: Adapter Framework code has been changed so that the invalid message is sent to the fault queue and does not stay in the request queue. |
|||
|
71388 |
2.6.0.4 |
JR22850 |
Symptom: When the “connector_manager –kill” command is used, Adapter Framework does not log messages. |
|
Resolution: Adapter Framework code has been changed so that a message like “The Connector <connector_name> was killed by <userid> at <date and time>” is logged in the connector_manager_<connector_name>.log |
|||
|
72509 |
2.6.0.4 |
JR22109 |
Symptom: RequestResponse data handler generates a ClassNotFoundException. The related logs are as follows: [Type: Trace]
[Mesg: :[9543] DoVerbFor cookie=Wed Jul 06 |
|
Resolution: Adapter Framework code has been changed so that the ClassNotFoundException does not occur. |
|||
|
74028 |
2.6.0.4 |
JR22747 |
Symptom: When JMS is the transport, the C++ adapter on the UNIX platform changes the child business object verb from CxBlank to CxIgnore. |
|
Resolution: Adapter Framework code has been changed so that the child business object verb does not change from CxBlank to CxIgnore. |
|||
|
74255 |
2.6.0.4 |
JR22851 |
Symptom: The adapterEnv.sh file needs to be corrected. The text “${WebSphere Business Integration A_JAR}” instead of “${WebSphere Business Integration A}” should be set to CWCLASSES. |
|
Resolution: The adapterEnv.sh file has been corrected. |
|||
|
2.6.0.4 |
(Enhancement) |
WebSphere Business Integration Adapter Framework backed-out message processing function has been enhanced. See the “Addition Discussion” section that follows for details. |
|
|
2.6.0.3 |
- |
There should have a value for an attribute which “Required Attribute” parameter is set true. If a required attribute does not contain a value, Adapter Framework will try to fill in the default value that is set in the business object specification. If there is not a default value in the business object specification, Adapter Framework will log an error message 17206 to the log file and processing the business object. |
|
|
61907 |
2.6.0.3 |
JR21718 |
Symptom: An unexpected exception occurs for CwConnector.dll. |
|
Resolution: Adapter Framework code has been changed so that no exception is generated for CwConnector.dll. |
|||
|
62206 |
2.6.0.3 |
JR21579 |
Symptom: When there are multiple child business objects, Adapter Framework does not check a child business object instance before setting the verb. |
|
Resolution: Adapter Framework code has been changed so that it compares the request business object with the response business object, and then determines which child business objects should be deleted, created, and updated. |
|||
|
60690 |
2.6.0.3 |
JR21987 |
Symptom: The adapter treats CxIgnore as an unsupported verb. |
|
Resolution: Adapter Framework code has been changed so that if the verb is CxIgnore, Adapter Framework sets it to null. |
|||
|
71559 |
2.6.0.3 |
JR21799 |
Symptom: Adapter Framework generates duplicate ObjectEventIDs as seen in the adapter log and trace file. |
|
Resolution: Adapter Framework code has been changed to use milliseconds instead of seconds as time granularity so that there will be no duplicate ObjectEventIDs. |
|||
|
71311 |
2.6.0.3 |
JR21772 |
Symptom: The adapter hangs if it gets specific application errors, such as the followings: · An error occurs due to an invalid search. · COM interface call error. Returned errorcode: 4096. · COM interface call error. Returned errorcode: 4175. |
|
Resolution: Adapter Framework code has been changed so that it provides an exception handling mechanism for the cases mentioned. |
|||
|
71515 |
2.6.0.3 |
JR21718 |
Symptom: After ‘q’ is pressed, Adapter Framework terminates the adapter, but does not suspend or stop the polling thread, which waits for the terminate method to finish. When the terminate method returns, the polling thread starts to run and JVM generates an unrecoverable error. |
|
Resolution: Adapter Framework code has been changed so that Adapter Framework stops the polling thread before terminating the adapter. |
|||
|
71399 |
2.6.0.3 |
JR21689 |
Symptom: The adapter does not stop polling events after the current MQ depth exceeds 90 percent of the maximum depth. |
|
Resolution: Adapter Framework code has been changed so that the adapter stops polling events after the current MQ depth exceeds 90 percent of the maximum depth. |
|||
|
72342 |
2.6.0.3 |
JR22046 |
Symptom: A Call Stack error exists in the log and trace file when the adapter is started. |
|
Resolution: Adapter Framework code has been changed to resolve the issue. |
|||
|
2.6.0.3 |
(Enhancement) |
MQ source does not stop polling when InterChange Server is stopped. |
|
|
70839 |
2.6.0.3 |
- |
When running natively on iSeries, the adapter ends gracefully through a terminate method instead of an iSeries-specific end signal. |
|
60381 |
2.6.0.3 |
JR21959 |
Symptom: When starting an adapter as a service, the cwservice command line is as follows: “cwservice -xi -mode=modeType -tCONNECTOR -cadapterServiceStartupScriptPath -sadapterName -iInterChangeServerName –tthreadModel” The “-t” parameter appears two times, the second one overwrites the first one. |
|
Resolution: Adapter Framework code has been changed so that there is no –t<Threadingmodel> parameter. |
|||
|
71857 |
2.6.0.3 |
JR21842 |
Symptom: When InterChange Server is the broker, synchronous scenarios do not work. |
|
Resolution: Adapter Framework code has been changed so that after the adapter sends a business object using synchronous mode, it can receive a response. |
|||
|
70953 |
2.6.0.3 |
JP21630 |
Symptom: When the business object schema and XML sequence is different, the business object is still processed and passed to the application. |
|
Resolution: Adapter Framework code has been changed so that the error message “Invalid BO’s attribute order” is logged and the current business object processing is stopped. |
|||
|
70841 |
2.6.0.3 |
JR21967 |
Symptom: After Adapter Framework fails to send a business object due to an out-of-memory exception, it cannot send the subsequent business objects although it can get them from the application. |
|
Resolution: Adapter Framework code has been changed so that it can continue to send the subsequent business objects even after receiving an out-of-memory exception when sending a business object. |
|||
|
70797 |
2.6.0.3 |
JR21986 |
Symptom: When JMS is the transport, if JMS type is invalid, the following error occurs. [MsgID: 9065][Mesg: The received message contains invalid JMS type. This message is ignored....] |
|
Resolution: Adapter Framework code has been changed so that if JMS type is invalid, the message “The received message contains invalid JMS type, this message will be sent to the fault queue.” is logged and the invalid message is sent to the fault queue. |
|||
|
66754 |
2.6.0.3 |
JR21962 |
Symptom: When the adapter is restarting, the following message occurs, but the cfg.bak file cannot be found in the specified directory. [System: Server] [Thread: wbia_main(#1378201390)] [Type: Info] [MsgID: 165] [Mesg: Updating configuration file “/wbi/WebSphereAdapters/connectors/*Connector.cfg”. The old configuration will be copied to “/wbi/WebSphereAdapters/connectors/*Connector.cfg.bak”.] |
|
Resolution: Adapter Framework code has been changed so that if the version of InterChange Server and the version in the connector configuration file do not match, a cfg.bak file will exist, but message 165 will not be logged. |
|||
|
66616 |
2.6.0.3 |
JR21352 |
Symptom: When an RFC_ERROR_SYSTEM_FAILURE error occurs, the adapter cannot restart. The related errors are as follows: [System: ConnectorAgent] [SS: SAPTJGSCMTestConnector] [Thread: Restart Thread - SAPTJGSCMTestConnector- retry #1 (#*)] [Type: Trace] [Mesg: :SetupTransport waiting for agent object to be exported] |
|
Resolution: Adapter Framework code has been changed so that when an RFC_ERROR_SYSTEM_FAILURE error occurs, the message “SetupTransport waiting for agent object to be exported” is logged and the adapter can restart. |
|||
|
70895 |
2.6.0.3 |
JR21958 |
Symptom: The following message is not clear. [MsgID: 190] [Mesg: While stopping a thread pool (QL.WBIA.JMSDS2.IQ2.PUBJmsThreadPool), the thread (Restart Thread - jmsds2.iq2Connector- retry #1) does not belong to the ThreadPool class and cannot be stopped. It is possible that the thread is a user self-generated thread. Please check that.] |
|
Resolution: Adapter Framework code has been changed so that the following message is now logged. [MsgID: 190] [Mesg: While stopping a thread pool (QL.WBIA.JMSDS2.IQ2.PUBJmsThreadPool), the thread (Restart Thread - jmsds2.iq2Connector- retry #1) does not belong to the ThreadPool class and cannot be stopped.]. “Restart Thread - jmsds2.iq2Connector- retry #1” has already explained the functionality of the thread. |
|||
|
66664 |
2.6.0.3 |
JR21414 |
Symptom: A lot of e-mails about CORBA errors cause a flooding of the network and mailbox. |
|
Resolution: Adapter Framework code has been changed so that some e-mails, but not too many, about CORBA errors are sent to the mailbox. |
|||
|
07633 |
2.6.0.3 |
(Enhancement) |
DataHandler base class provides a mechanism for displaying the version of the data handler. |
|
35953 |
2.6.0.3 |
(Enhancement) |
DataHandler base class provides PrintStackTrace functionality. |
|
25927 |
2.6.0.3 |
(Enhancement) |
The property org.apache.xerses.xni.parser.XMLParserConfiguration must be set when using Xerces 2 as the parser. |
|
33724 |
2.6.0.3 |
(Enhancement) |
An adapter can be run as a Windows service in stand-alone mode. |
|
37142 |
2.6.0.3 |
(Enhancement) |
The adapter has its own message file <AdapterName>Connector.txt. But for some reason, perhaps an incorrect path, Adapter Framework cannot locate this file. When this occurs, Adapter Framework defaults to WBIAdapterRuntime.txt as the adapter message file, which might lead to misleading error messages. Therefore if the adapter message file is missing, a message will be printed to indicate that the adapter message file cannot be located and shut down. |
|
28564 |
2.6.0.3 |
(Enhancement) |
Adapter Framework will print its version number in the log file, not only in the trace file. |
|
61985 |
2.6.0.2 |
|
Symptom: The command “connector_manager –stopgraceful” does not work as documented. When running this command, the following error occurs. Usage: StopAgent -aagentName -sserverName |
|
Resolution: Adapter Framework code has been changed so that “connector_manager –stopgraceful” is supported and can work as documented. |
|||
|
44949 |
2.6.0.2 |
|
Symptom: Adapter run time cannot accept a configuration file that has a version of 4.4.0. |
|
Resolution: Adapter Framework code has been changed so that it can accept a CFG file that has a version of 4.4.0. |
|||
|
60685 |
2.6.0.2 |
|
Symptom: When JMS is the transport, the
adapter fails to convert the XML message with Message
Domain as “MRM” to a business object in the request
queue. While it succeeds to convert the XML message with Message Domain as
“mrm” to a business object. |
|
Resolution: Adapter Framework code has been changed so that it can recognize “MRM” also. |
|||
|
43926 |
2.6.0.2 |
|
Symptom: CHANNEL and PORT are required in the adapter configuration file when using a local MQ binding. |
|
Resolution: Adapter Framework code has been changed so that when using a local MQ binding, CHANNEL and PORT are not required. |
|||
|
2.6.0.2 |
|
Symptom: The adapter fails to process the XML message without a namespace when JMS is the transport. |
|
|
Resolution: Adapter Framework code has been changed so that it supports the XML message with no namespace. See the “Additional discussion” section below for details. |
|||
|
61281 |
2.6.0.2 |
|
Symptom: The adapter generates an exception when sending invalid messages to the fault queue when JMS is the transport. The related logs are as follows: [MsgID: 84017] [Mesg: Unable to deliver the message to the Queue. Reason: java.lang.ClassCastException: com.ibm.jms.JMSBytesMessage] [MsgID: 9061] [Mesg: Caught JMS error. Reason: javax.jms.JMSException: java.lang.ClassCastException: com.ibm.jms.JMSBytesMessage.] |
|
Resolution: Adapter Framework code has been changed so that the adapter succeeds in sending invalid messages to the fault queue. |
|||
|
61176 |
2.6.0.2 |
|
Symptom: When JMS is the transport, a NullPointerException occurs when the synchronous request times out to get the response. The related logs are as follows: [Type: Trace] [Mesg: :Send.sendMessageSync got response --null] java.lang.NullPointerException |
|
Resolution: Adapter Framework code has been changed so that Adapter Framework does not generate a NullPointerException. |
|||
|
44904 |
2.6.0.2 |
|
Symptom: The JMS transported adapter cannot get the response in the Sync model. The related logs are as follows: [Type: Error] [MsgID: 80] [Mesg: Unable to construct BusinessObject from network representation of businessobject definition: . Incoming network representation: .] [Type: Error] [MsgID: 81] [Mesg: An unexpected exception occurred during the construction of a business object. Reason: java.lang.NullPointerException] |
|
Resolution: Adapter Framework code has been changed so that the JMS transported adapter can get a response in the Sync model. |
|||
|
44363 |
2.6.0.1 |
|
Symptom: When JMS is the transport, the following actions are taken depending on the value of DuplicateEventElimination: (1) If the property DuplicateEventElimination (DEE) is set to TRUE, the WSDL binding information is not added to the JMS message by Adapter Framework. (2) If the property DuplicateEventElimination (DEE) is set to FALSE, the WSDL binding information is added to the JMS message by Adapter Framework. Actually, Adapter Framework should add
the WSDL binding information regardless of the value of |
|
Resolution: Adapter Framework code has
been changed so that Adapter Framework adds the WSDL binding information to
the JMS message regardless of the value of |
|||
|
44635 |
2.6.0.1 |
|
Symptom: In MAIN_SINGLE_THREADED mode, an application call is made from the main thread as a single thread. The problem is that the event polling thread cannot be started correctly. |
|
Resolution: Adapter Framework code has been changed so that the event polling thread can be started correctly in MAIN_SINGLE_THREADED mode. |
|||
|
2.6.0.1 |
(Enhancement) |
Adapter Framework provides built-in message audit and tracking functionality. |
|
|
2.6.0.1 |
(Enhancement) |
When JMS is the transport, Adapter Framework cannot handle security exits used for WebSphere MQ queue managers. |
|
|
59346 |
2.6.0.1 |
|
Symptom: The XML DataHandler succeeds in transforming one XML document into a set of business objects. However, the business object output by Adapter Framework is not acceptable by the adapter. |
|
Resolution: Adapter Framework code has been changed so that the business object output is normal. |
|||
|
44746 |
2.6.0.1 |
|
Symptom: In the log file and trace file, the Adapter Framework version is the InterChange Server version. Because Adapter Framework is separate from InterChange Server, the Adapter Framework version in the log and trace file should be that of the Adapter Framework version itself. |
|
Resolution: Adapter Framework code has been changed so that: · The version “4.3.0.0” is replaced with “2.6.0.1”. · The version “4.3.0” is replaced with “2.6.0”. · If InterChange Server is the broker and IDL is the transport, Adapter Framework can get the version of InterChange Server and print it. |
WebSphere Business Integration Toolset changes
|
Problem tracking numbers |
As of Version |
Problem |
|
69280 |
4.3.0.5 |
Symptom: When JMS is the transport, test
connector generates a NullPointerException when processing request message
with ObjectEventId as CxIgnore. |
|
Resolution: WebSphere Business
Integration tools code has been changed so that it can process request
message successfully. |
||
|
35232 |
4.3.0.2 |
A problem occurs with trace/log files when “Locale” is set to French. |
|
70818 |
4.3.0.2 |
4.2.2.5 Connector Configurator flattens hierarchical configuration files. |
|
71926 |
4.3.0.2 |
Connector Designer 4225 issues with HTTP Connector 1.0. |
|
45463 |
4.3.0.2 |
Connector Configurator will not load customer’s template for the 2nd time. |
|
73167 |
4.3.0.2 |
HCG_BIDI: Connector Configurator does not support WebSphere Application Server broker. |
|
72221 |
4.3.0.2 |
An appended connector-specific property disappeared after 4.3.0.1 upgrade. |
|
73888 |
4.3.0.2 |
Connector Configurator is slow when loading a business object. |
|
66879 |
4.3.0.2 |
Test Connector has problem to display correctly when the Child business object is <CxIgrone>. |
|
44484 |
4.3.0.1 |
A new connector property is needed to explicitly turn the business object auditing message on and off. |
|
60329 |
4.3.0.1 |
Slow InterChange Server bring-up time causes auto-start adapters to fail on start-up |
|
45637 |
4.3.0.1 |
Connector Configurator should validate the values of connector-specific properties. |
|
44774 |
4.3.0.1 |
A “no” option should be added to the property XMLNameSpaceFormat |
|
45148 |
4.3.0.1 |
The description of the standard property SynchronousRequestTimeout is incorrect. |
|
44800 |
4.3.0.1 |
The version of the adapter is incorrectly displayed in Adapter Monitor. |
|
59069 |
4.3.0.1 |
zh_tw_pre-TVT_english string has problem on Load Adapter |
|
45586 |
4.3.0.1 |
“adminmgr” does not work normally |
|
62182 |
4.3.0.1 |
Business Object Designer version 4.3 has slow performance. |
|
61342 |
4.3.0.1 |
ICL projects disappear from System Manager |
|
60726 |
4.3.0.1 |
The Log Viewer cannot be launched by specifying the log file as an argument |
|
61190 |
4.3.0.1 |
Business-object level application-specific information disappears when Business Object Designer is restarted. |
Prerequisites
You must have WebSphere Business Integration Adapter Framework version 2.6.0 and/or Toolset installed in order to install this fix pack.
To install this fix pack on HP IA 64 platform, you must have WebSphere Business Integration Adapter Framework version 2.6.0.10 installed.
Installing the fix pack
1. Ensure that Adapter Framework 2.6.0 (or Adapter Framework 2.6.0.10 for HP IA 64 platform) is installed and note the installation directory.
2. Download the fix pack zip or tar file and unzip or untar the file to a local directory.
3. Run the fix pack Installer corresponding to your operating system:
- Windows – setupwin32.exe*
- Linux – setupLinux.bin
- AIX– setupAIX.bin
- HP UX – setupHP.bin
- HP IA64 - setupHPIA64.bin
- Solaris – setupSolarisSapc.bin
Refer to the Installer technical notes for instructions for installing Adapter Framework 2.6.0.x on the following platforms:
· AIX 5.3.
Refer to the Installer technical notes for instructions for installing Adapter Framework 2.6.0.x on the following platforms:
· AIX 6.1.
Refer to the Installer technical notes for instructions on installing Adapter Framework 2.6.0.x on the following platforms:
· SUSE 9.0 and SUSE10
Refer to the Installer technical notes for instructions on installing Adapter Framework 2.6.0.x on the following platforms:
· HP_UX 11i V2 and HP_UX 11i V3
Refer to the Installer technical notes for instructions on installing Adapter Framework 2.6.0.5 or later on the following platforms:
· Solaris10
Refer to the Installer technical notes for instructions on installing Adapter Framework 2.6.0.5 or later on the following platforms:
· Red
Hat
Note: You can record the options you select when installing Adapter Framework for subsequent silent installations. For more information, see “Installing silently.”
*For Windows Vista and Windows Server 2008, enter “setupwin32.exe –W unsupportedPlatform.active=false” on the command line.
4. Follow the Installer wizard prompts to complete the installation of the Adapter Framework fix pack. Note that the Installer will end or not proceed if any of the following occurs:
You decline the license agreement
You are installing on a Windows platform and you do not have administrative privileges, or if your user login exceeds 20 characters
You provide a destination directory name that is different from the installation directory for Adapter Framework 2.6.0 (or Adapter Framework 2.6.0.10 for HP IA 64 platform)
You do not provide sufficient free disk space (On Windows you need 300 Mb; on all UNIX platforms except HP IA64 you need 150 Mb; on HP IA 64 you need 400 Mb)
5. During fix pack installation, you may be prompted to overwrite existing files. To successfully install the fix pack, you must select Yes to All.
Windows registry entries
After installation on Windows systems, the fix pack Installer updates a registry folder under HKEY_LOCAL_MACHINE\SOFTWARE\IBM\WebSphere Business Integration.
Installing silently
1. Launch the fix pack installer by using the following command:
<install launcher executable> -options-record<install_settings_filename>
For example, on Windows systems, enter
setupwin32.exe –options-record mySettings.txt
2. Edit the file containing your settings (for example, mySettings.txt) by adding the following lines to the end of the file:
–G replaceExistingResponse=“yesToAll ”
–G replaceNewerResponse=“yesToAll ”
–G createDirectoryResponse=“yes ”
–G removeExistingResponse=“yesToAll ”
–G removeModifiedResponse=“yesToAll ”
For an example, see the settings.txt file bundled with the fix pack installer.
3. To silently install, launch the fix pack installer as follows:
<Install launcher executable> -silent-options<install_settings_filename>
For example, enter:
setupwin32.exe –silent-options mySettings.txt
Adapter Framework runtime files installed by this fix pack:
|
Platform [Win32;Solaris;AIX;HP_UX;HP_IA;Linux] |
Add / Replace / Remove |
File (Starting from WebSphere Business Integration Adapters installation directory) |
|
Win32:Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
wbiart/wbiart.jar |
|
Win32:Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
messages/WBIAdapterRuntime.txt |
|
Win32:Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
lib/WBIA.jar |
| Win32:Solaris:AIX:HP_UX:Linux | Replace | connectors/messages/CwConnectorAgentMessages.txt |
| Win32:Solaris:AIX:HP_UX:Linux | Replace | 5724G92J020600.sys |
|
Win32:Solaris:AIX:HP_UX:Linux |
Replace |
license.txt |
|
Win32:Solaris:AIX:HP_UX:Linux |
Replace |
notices.txt |
|
Win32:Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
DataHandlers/CwDataHandler.jar |
|
Win32:Solaris:AIX:HP_UX:Linux |
Replace |
lib/cei/commons-jxpath.jar,
lib/cei/events-client.jar, lib/cei/events-consumer.jar |
|
Win32:Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
lib/jmsmgr.jar |
|
Win32:Solaris:AIX:HP_UX:Linux |
Remove |
lib/xwbase.jar, lib/xworacle.jar, lib/xwsqlserver.jar, lib/xwutil.jar |
|
Win32:Solaris:AIX:HP_UX:Linux |
Add |
lib/aspectj/aspectjrt.jar,
lib/ffdc/ffdc.jar, lib/ffdc/ffdcSupport.jar, lib/ffdc/wsexception.jar,
lib/ffdc/analysisEngine.jar, lib/ffdc/properties/ffdcStart.properties,
lib/ffdc/properties/ffdcRun.properties,
lib/ffdc/properties/ffdcStop.properties |
|
Win32 |
Replace |
bin/cwservice.exe |
|
Win32 |
Replace |
bin/start_adapter.bat |
|
Win32 |
Replace |
bin/start_connector.bat |
|
Win32 |
Replace |
bin/CWSharedRuntimeEnv.bat |
|
Win32 |
Replace |
bin/ODKRuntimeEnv.bat |
|
Win32 |
Replace |
bin/CWConnEnv.bat |
|
Win32 |
Replace |
bin/adminmgr.bat |
|
Win32 |
Replace |
bin/start_VTConnector.bat |
|
Solaris:AIX:HP_UX:Linux |
Replace |
bin/adapterEnv.sh |
|
Solaris:AIX:HP_UX:Linux |
Replace |
bin/start_adapter.sh |
|
Solaris:AIX:HP_UX:HP_IA:Linux |
Replace |
bin/connector_manager |
|
Solaris:AIX:HP_UX:Linux |
Replace |
bin/CWSharedEnv.sh |
|
Solaris:AIX:HP_UX:Linux |
Replace |
bin/ODKRuntimeEnv.sh |
|
Solaris:AIX:HP_UX:Linux |
Replace |
bin/adminmgr.sh |
|
Win32:Solaris:AIX:HP_UX:Linux |
Replace |
AdapterJRE/ (whole Dir) |
|
Win32:Solaris:AIX:HP_UX:Linux |
Add/Replace |
jdk/ (whole Dir) |
|
Win32 |
Replace |
bin/CwConnector.dll |
|
Solaris |
Replace |
lib/libCwConnector.so |
Toolset files installed by this fix pack – For Windows installations only:
|
Platform [Win32;Solaris;AIX;HP_UX;Linux] |
Add / Replace / Remove |
File (Starting from the WebSphere Business Integration Adapters installation directory) |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_chs_chn.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_cht_twn.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_deu_deu.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_esn_esp.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_fra_fra.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_ita_ita.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_jpn_jpn.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_kor_kor.xml |
|
Win32 |
Replace |
bin/Data/Std/StdConnProps_ptb_bra.xml |
|
Win32 |
Replace |
bin/BusObjDesigner.exe |
|
Win32 |
Replace |
bin/ConnDesigner.exe |
|
Win32 |
Replace |
bin/LogViewer.exe |
|
Win32 |
Replace |
bin/enu_usa/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/chs_chn/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/fra_fra/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/jpn_jpn/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/cht_twn/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/ita_ita/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/deu_deu/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/esn_esp/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/ptb_bra/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/kor_kor/ConnDesignerRes.dll |
|
Win32 |
Replace |
bin/CxCommon2.dll |
|
Win32 |
Replace |
bin/ServerConnection.dll |
|
Win32 |
Replace |
bin/XRmi.dll |
|
Win32 |
Replace |
bin/SMInterface.dll |
|
Win32 |
Replace |
lib/CrossWorlds.jar |
|
Win32 |
Replace |
lib/datamanager.jar |
|
Win32 |
Replace |
lib/vtc.jar |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.csm |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.common |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.codeGen |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.ae.settings |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.cwconverter |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.datamanager |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.itools.wsdlgen |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.adaptermonitor |
|
Win32 |
Replace |
Tools/eclipse/plugins/com.ibm.btools.bdm |
Refer to technote 1271602 for details.
1. To enable the MQ Security Exit (CR58998) for the adapter:
- Launch Connector Configurator and then set the type of DeliveryTransport to “JMS.”
- Insert the class name MQ Security Exit as the last element by using this format: “queue_manager_name:channel_name:host_name:port_number:mq_security_exit_name”
- Put the class MQ Security Exit in the library path or extend the path of the adapter.
- When the adapter is launched, the MQ Security Exit function will be invoked. If MQ Security Exit cannot be found, this feature will be treated as disabled.
Note: Because MQ Security Exit applies to queue manager, products that do not support the MQ Security Exit feature should not use the same queue manager as Adapter Framework.
2. To use the BO trace (CR59316) feature, enable the “BOTrace” property:
- If the patch for Connector Configurator is not available and you are using InterChange Server or a stand-alone broker, you can use the “-x” parameter in the agent start script to configure the “BOTrace” property, for example “-xBOTrace=Full.”
- If the patch for Connector Configurator patch is available, you can configure the “BOTrace” property in Connector Configurator.
The “BOTrace” values are as follows:
When “AgentTraceLevel” is 5, the property “BOTrace” is not available.
When “AgentTraceLevel” is lower than 5:
- Set “BOTrace” to “Full” to dump all the business object attributes.
- Set “BOTrace” to “Keys” to dump only the business object keys.
- Set “BOTrace” to “None” to disable business object attribute dumping.
3. To support no namespace BO (CR37590)
If the “no” option is not available for the property XMLNameSpaceFormat in Connector Configurator, you should edit the Connector Configuration file manually, as described below. (It should be added in the WebSphere Business Integration Toolset version 4.3.0.1)
- Open the Connector Configuration file in a text editor.
- Locate the “XMLNameSpaceFormat” property by searching. Here is an example:
<tns:property>
<tns:name>XMLNameSpaceFormat</tns:name>
<tns:type>String</tns:type>
<tns:subtype></tns:subtype>
<tns:value xml:space="preserve">short</tns:value>
<tns:description>Supports short and long name spaces in the business object XML doc on the wire.</tns:description>
- Replace the value “short” or “long” by “no”, as follows:
<tns:property>
<tns:name>XMLNameSpaceFormat</tns:name>
<tns:type>String</tns:type>
<tns:subtype></tns:subtype>
<tns:value xml:space="preserve">no</tns:value>
<tns:description>Supports short and long namespaces in the business object XML doc on the wire.</tns:description>
4. AIX 5.3 support:
WebSphere Business Integration Adapter Framework has supported AIX 5.3 since version 2.6.0.2.
5. To enable Adapter Monitor to support the MQ Security Exit feature (CR58998):
- Put the class MQ Security Exit into the directory <Adapter Framework Installer>\Tools\eclipse\plugins\com.ibm.btools.adaptermonitor\lib.
- Modify the Adapter Monitor’s plug-in manifest file plugin.xml which resides in the directory <Adapter Framework Installer>\Tools\eclipse\plugins\com.ibm.btools.adaptermonitor. In the <runtime> segment, add a library name entry for MQ Security Exit.
For example, suppose MyMQExit.zip includes this class. Add the following library name entry to the <runtime> segment of plugin.xml.
<library name="lib\MyMQExit.zip"/>
6. To enable the enhancement (CR66866) with MQ DeliveryTransport for the adapter:
A new command line parameter “WhenServerAbsent” is incorporated to indicate whether the adapter with MQ transport should pause or shut down when InterChange Server is stopped. If you are using InterChange Server, you can use the “-x” parameter in the agent start script to configure the “WhenServerAbsent” property, for example:
-xWhenServerAbsent=shutdown //Adapter with MQ transport will shut down if InterChange Server is not available
-xWhenServerAbsent=pause //Adapter with MQ transport will pause if InterChange Server is not available
If the option is not set, the adapter will keep polling by default.
7. Adapter Framework backed-out message processing function is enhanced (CR72029):
Command line parameter “CheckRedelivered” is incorporated to indicate Adapter Framework how to handle redelivered message.
-xCheckRedelivered=<Threshold>
<Threshold> should be zero or a positive number. Adapter Framework will try to process messages up to that number of times before putting them in the fault queue.
- The default value is 1, which makes Adapter Framework behave as it did before this parameter was included.
- A value of zero means Adapter Framework will try to process the redelivered message instead of putting it in the fault queue. The number of times it has been redelivered will be written to trace file.
- If the value provided is invalid, Adapter Framework uses the default value of 1.
8. Messages from WMQI broker that has no namespace is supported (CR72313, CR73664):
Command line parameter “AFNamespaceAware” is incorporated to indicate whether the XML parser will check the namespace.
-xAFNamespaceAware=true or is not set //Default. Working with xml XML message that has the namespace
-xAFNamespaceAware=false //Working with XML message that does not have the namespace.
Note:
Adapter Framework will not support XML messages with namespace and without namespace at same time.
9. To resolve the problem in CR71557, WebSphere InterChange Server V4.3 need to apply fix pack 2 or later version.
10. When Websphere Business Integration Message Broker V6 is the broker, follow the documentation for WebSphere Business Integration Message Broker V5.
- If you are migrating from WebSphere MQ Integrator Broker 2.1 by using mqsimigratemsgflows.exe, add the parameter “-data” to locate the workspace as shown in thi example, mqsimigratemsgflows.exe -p <project name> –d “<your message flow files directory>” -data <WebSphere Business Integration MB6 directory>\eclipse\workspace
- If you are migrating from WebSphere Business Integration MQ V5, note that the queue manager listener default port is 2414 in WebSphere Business Integration MB V6, but it is 1414 in WebSphere Business Integration MB V5. You may need to change the default port to 1414 to be consistent with the queue manager and the adapter configuration.
The Required property specifies whether a business object attribute must contain a value. If a particular attribute in the business object must contain a value to be able to process the business object data, set the Required property for the attribute to “true.” Before the fix for CR45674 is applied, a business object can be processed even when its required attributes do not have values. This incorrect behavior was fixed in the fix pack for CR45674. If a Required attribute does not contain a value, Adapter Framework will try to fill in the default value that is set in the business object specification. If there is not a default value in the business object specification, Adapter Framework will log an error message 17206 to the log file and processing the business object. For a Required attribute, a value should be set by an adapter or a default value should be set in the business object specification.
12． To provide a new restart/retry behavior, a new -x parameter has been added.(CR84135)
-xMaxStartRetry=-1 (or any negative integer)
The adapter retries to connect to the controller until it succeeds.
-xMaxStartRetry=n (n>0)
The adapter retries n times to connect to the controller. The interval between attempts is one minute.
-xMaxStartRetry=0
The adapter does not retry to connect to the controller.
Note: If the option is not set, the adapter retries to connect to the controller for a maximum of three times.
13． CR27882
Library packages xwbase.jar, xworacle.jar, xwsqlserver.jar and xwutil.jar have expired under the <fix_pack_install_location>/lib directory. They have been removed because these embedded JDBC drivers are not necessary for Adapter Framework and are not required for Adapter for JDBC. To configure JDBC drivers for Adapter for JDBC, see the following link.
14． CR84583
In this enhancement, Adapter Framework waits for all working threads to finish processing before shutting down. A timeout value for shutting down will be set when the adapter is started. If any working threads do not finish in the timeout timeframe, Adapter Framework will stop them. All messages that are being processed will be put into the fault queue after the adapter restarts. The –x parameter sets the timeout value on the adapter when starting rather than when stopping. The unit of this value is second.
- To start the adapter with a timeout parameter value in minutes
connector_manager –start <connect_name> -x"ShutdownTimeout=5"
- To shut down the adapter
connector_manager -stop|-stopgraceful
15． CR69014
If WebSphere Process Server or WebSphere Application Server is the broker, the NULL value will be converted to CxIgnore when converting a business object to an XML message. If you still want NULL to be converted to Blank when WebSphere Application Server is the broker, you can use a new –x parameter.
Add the parameter –xNullToBlank to the adapter starting command line
-xNullToBlank=false or is not set //Default. Convert NULL to CxIgnore
-xNullToBlank=true //Convert NULL to Blank
16． CR88143
This fix only supports Adapter for Clarify which cannot shutdown completely on the Windows 2000 platform. Other adapters are not affected.
17． CR87520
If using MQ as a remote queue, the flow control feature is disabled because the remote queue does not support inquiring about queue depth. The flow control feature stops the agent from polling for events after the current MQ depth exceeds 90 percent of the maximum depth. This feature was first included in Adapter Framework V2.6.0.3 by CR73199.
18． CR86571
When a child business object verb is not set (if it is either null or “”), Adapter Framework can optionally populate the child business object verb according to its nearest parent business object verb, which is set when sending it to the delivery queue. A new –x parameter should be used.
Add parameter –xPopChildVerb to the adapter starting command line
-xPopChildVerb=false or is not set //Default. It is not populated the verb if it is not set
-xPopChildVerb=true //Populate the verb if it is not set according to its nearest parent BO verb which is set
19． CR69146
Refer to technote 1272049 for details.
20． CR88502
You can set a customized JVM parameter when starting the adapter on the UNIX platform as of Adapter Framework fix pack 2.6.0.9.
In the start_<connector>.sh file (for example, start_SAP.sh), add the following lines before “exec ${CROSSWORLDS}/bin/start_adapter. sh…”
JVM_ARGS=“-Xms128m -Xmx712m”
export JVM_ARGS
Notes:
1. The text –Xms128m means to set the initial Java heap size to 128m, -Xmx712m means to set maximum Java Heap Size to 712m. You can set the parameter to other values.
2. There is no blank between “JVM_ARGS” and ‘=’.
3. If the customized JVM parameter is not set, then the default value will be used. The default value is “-Xms256m –Xmx512m.”
4. There is no check for the customized JVM parameter. Ensure the customized value is correct and meaningful.
21． CR88205
To enable Adapter Framework local transaction with WebSphere MQ JMS, a new –x parameter has been introduced.
Add parameter -xAFJMSTransacted=true to the adapter starting command line
-xAFJMSTransacted=false or is not set //Default. JMS Transaction is disabled
-xAFJMSTransacted=true //JMS Transaction is enabled
When adapters work in a transacted state, more MQ log space is needed. Therefore, you must adjust log space size according to your specific situation when JMS transaction is enabled.
22． CR89146
To enable you to log customized time-related information on WebSphere Business Integration Adapter Framework synchronous calls, two new –x parameters have been introduced.
Add –xLogConsumeSync and –xLogClass to the adapter starting command line.
-xLogConsumeSync=false or doesn’t set //Default. Disable logging time related information on synchronous call
-xLogConsumeSync=true //Enable logging time related information on synchronous calls
-xLogClass=** //Specify the customized log class name used to log time related info. This parameter takes effect only when LogConsumeSync is set to “true.”
Customized log class needs to implement an interface named ConsumeSyncLog. The definition of ConsumeSyncLog.java in Adapter Framework is as follows:
**************************************************************************************
package AppSide_Connector;
import CxCommon.BusinessObject;
public interface ConsumeSyncLog {
public void log(long startTime, long
elapsedTime, String collabName, BusinessObject bo);
}
**************************************************************************************
Following is an example of source code of log class- Log.java.
**************************************************************************************
import AppSide_Connector.ConsumeSyncLog;
import CxCommon.BusinessObject;
public class Log implements ConsumeSyncLog
{
public void log(long startTime, long
elapsedTime, String collabName, BusinessObject bo) {
System.out.println("*******In
self-defined Log class*******");
System.out.println("ConsumeSync
startTime: "+ startTime+": " +new
java.text.SimpleDateFormat("yyyy-MM-dd k:m:s").format(new
java.util.Date(startTime)) +" elapsedTime: "+ elapsedTime +"
collabName: "+ collabName+" bo"+bo.dump());
}
}
**************************************************************************************
Note that the Log.java file needs to import the two classes shown below. AppSide_Connector.ConsumeSyncLog is the interface provided by Adapter Framework to allow customized logging for consumesync related time info. CxCommon.BusinessObject need be imported because the log method in the ConsumeSyncLog class has one parameter of CxCommon.BusinessObject type.
import
AppSide_Connector.ConsumeSyncLog;
import
CxCommon.BusinessObject;
Due to import of these two Adapter Framework classes, when compiling Log.java, you must ensure that the compiler can find the wbiart.jar file in the classpath or extdirs.
After compilation, customer need jar Log.class to Log.jar and include this JAR file in the classpath of the adapter start up script.
The following is a sample which uses Test Connector (VTC).
a. Change the directory to the one where the Log.java file is located. For example: E:\.
b. Check the path where the wbiart.jar file is located, for example: C:\IBM\WebSphereAdapters\wbiart.
c. Compile the Log.java file by using the command: javac -extdirs C:\IBM\WebSphereAdapters\wbiart Log.java
d. Run the command: jar cvf Log.jar Log.class
e. Revise the start_VTConnector.bat file by adding the Log.jar file in the classpath as shown below.
REM for Java classpath
set
CP=%VTC%;%JAVA_MAIL%;%ARCLASSES%;%CWORION%;%DATAHANDLER%;%BIDI%;%ARM_JNI%;%WASCLIENT_PROPERTY_PATH%;%FFDC%;E:\Log.jar
f. Start InterChange Server, then start VTC as the source with the following two parameters “-xLogConsumeSync=true -xLogClass=Log.” Start another adapter (use another VTC) as the destination and then use the source VTC to send a synchronous mesage.
After the destination VTC responds the message, you can see the following information printed in the source VTC console as implemented in the Log.java file.
*******In self-defined Log class*******
ConsumeSync startTime: 1184741240550:
2007-07-18
<Version = 3.0>
<EndHeader>
<StartBO:testBO>
BusinessObject = testBO
Verb = Create
Locale
= en
ID
= 2
NAME = 2
GENERAL = 1
NIANLING = 1
ObjectEventId = myVtcConnector_1184741240534_1
<EndBO:testBO>
23． CR89152
So that you can optionally write AP queue messages as non-persistent, a new –x parameter has been introduced.
Add –xNonPersistMQMsg to the adapter starting command line.
–xNonPersistMQMsg=false or is not set //Default. Write the AP queue message as persistent
–xNonPersistMQMsg=true //Write AP queue message as non-persistent
24． CR91674
How to use ReplaceCxIgnore property:
Users can set property "ReplaceCxIgnore" in adapter starting parameter or configuration file.
1. If ReplaceCxIgnore=true, when the attribute value starts with '\0', Adapter Framewok will convert it into CxIgnore.
2. If ReplaceCxIgnore=false, when the attribute value starts with '\0', Adapter Framewok won’t convert it into CxIgnore.
3. If ReplaceCxIgnore is not set, Adapter Framewok will set it to true.
4. If ReplaceCxIgnore is set with other values, one error message[84034] with " Property "ReplaceCxIgnore" value is invalid since it is set to ***. It can be only set to true or false. If it is not set, adapter will take true as default" will be logged.
25． For the summary of all new properties added from Adapter Framework V2.6.0.1, refer to technote 1269240.
You can uninstall this fix pack by using an interactive (graphical) tool or silently (by using scripts).
To launch the graphical uninstaller on Windows platforms, take one of the following actions:
- Choose Start>Settings>Control Panel, double-click Add/Remove Programs, and specify the Adapter Framework
- Run the following executable file:
<fix_pack_install_location>_uninst_ Adapter Framework\uninstaller.bin
To launch the graphical uninstaller on UNIX or Linux platforms:
- Run the following executable file:
<fix_pack_install_location>_uninst_ Adapter Framework/uninstaller.bin
After you launch the uninstaller, follow and respond to the prompts presented by the wizard.
To launch a silent uninstallation on Windows platforms:
- Run the following executable file:
<fix_pack_install_location>_uninst_ Adapter Framework\uninstaller.exe –silent
To launch a silent uninstallation on UNIX or Linux platforms:
Run the following executable file:
<fix_pack_install_location>_uninst_Adapter Framework/uninstaller.bin –silent
© 2008 IBM Corporation. Proprietary and Confidential. All Rights Reserved.