Resolving problems when starting resources

Use the advice given here to help you resolve common problems that can arise when you start resources other than an integration node.

About this task

To learn how to resolve common problems when starting an integration node, see Resolving problems when starting an integration node.

For advice about specific problems that can occur when you start resources, see the following sections.

Resources terminate during startup

Procedure

  • Windows platformScenario: The following error message is displayed when you start an integration node on Windows: 
    ServiceName - DLL initialization failure Initialization of the
dynamic link library c:\windows\system32\user32.dll failed.
The process is terminating abnormally.
  • Explanation: This error is issued by Windows when it fails to start a service because it has insufficient storage.
  • Solution: This error is an operating system problem. Information about how to recover from this problem is available in the Microsoft Developer Network (MSDN). You can access MSDN on the Web at http://msdn.microsoft.com.

Resources hang at startup on Windows

Procedure

  • Windows platformScenario: You try to start the integration node on Windows, but nothing happens in the Event log to show that a connection has started.
  • Explanation: This problem is typically caused by processes having only one thread. To see if this is the cause, check the Windows Task Manager. If either of the processes bipconfigmgr.exe or dataflowengine.exe has started, check the number of threads owned by the process. If the process has only one thread, you are likely to encounter this problem.
  • Solution:
    1. Shut down the integration node using the mqsistop command and end the process from within the Task Manager.
    2. From the Windows Start button, click Settings > Control Panel
    3. Double-click Administrative Tools
    4. Double-click Services to open the Services window. From the list of available services, locate and right-click the integration node resource that you want to start (the service name begins with IBM® Integration Bus component). Click Properties from the menu.
    5. Make a note of the This Account setting. Contact the system administrator to obtain the password associated with This Account, because these settings are lost when you change values.
    6. Select System Account as the Log On As option, and select Allow Service to Interact with Desktop. These selections allow you to see any hidden dialog messages. Click OK to accept the changes.
    7. Restart the resource that is failing and report any subsequent error messages and dialog box messages to your IBM Service Representative.
    8. When your IBM Service Representative has resolved this problem for you, make sure that you restore the This Account, Password, and Confirm Password entries to the values that you used when you created the integration node.

Error message BIP8048 is issued when you start a component

Procedure

  • Scenario: Error message BIP8048 is issued when you start a component.
  • Explanation: This message indicates that WebSphere® MQ is not responding as expected when it tries to start the queue manager. This problem might be because of the following reasons:
    • Linux platformUNIX platformThe strmqm executable file is not present on Linux® or UNIX systems.
    • Windows platformThe amqmdain executable file is not present on Windows
  • Solution: Check that your WebSphere MQ installation is fully functional:
    • On Windows, start the "IBM MQSeries®" service.
    • On UNIX or Linux, issue the strmqm command to start the queue manager that is associated with this component.

    If the check fails, your WebSphere MQ installation is incomplete. This error occurs typically because you have previously installed WebSphere Application Server, which installs an embedded WebSphere MQ component that does not support IBM Integration Bus.

    Uninstall WebSphere Application Server, then install the full WebSphere MQ product that is provided with IBM Integration Bus.

A "Not Found" error is issued when you click a link to a specific sample

Procedure

  • Scenario: You see a "Not Found" error when you click a link to a specific sample, indicating that a URL is invalid.
  • Explanation: You can view sample applications only when you use the product documentation that is integrated with the IBM Integration Toolkit. If you are viewing a stand-alone or online product documentation, you cannot access these resources.
  • Solution: If you want to access samples, ensure that you are viewing the product documentation from within the IBM Integration Toolkit.

Error message BIP0832 is issued on startup

Procedure

  • Scenario: The following error message is displayed on startup:

    BIP0832E: A class java.io.FileNotFoundException exception occurred which reported the following message: [filepath] (The process cannot access the file because it is being used by another process). Resolve the reason of error and try again.

  • Explanation: An invalid WebSphere MQ Java™ Client trace output file has been specified on the Enqueue preferences screen.
  • Solution:
    1. Open the Enqueue preferences screen by clicking Windows > Preferences, then clicking Enqueue on the left.
    2. In the To file field, specify a valid output file (one that is not read-only or already in use).

Your integration servers restart repeatedly

Procedure

  • Scenario: Your integration servers restart repeatedly. The system log might show an error, such as BIP2060.
  • Explanation: The problem might be caused by:
    • The integration node environment variables incorrectly defined
    • Incorrect Loadable Implementation Library directory permissions
    • Incorrect database permissions
    • Invalid user-written LILs
  • Solution: Check:

You cannot tell whether startup is complete on z/OS

Procedure

  • Scenario: You cannot tell whether startup has completed on your z/OS system.
  • Solution: To determine if startup is complete:
    1. Check the messages in the system log. The following example shows a system log entry for a startup of an integration node with one integration server: 
      S STU3053
$HASP100 STU3053  ON STCINRDR
IEF695I START STU3053  WITH JOBNAME STU3053  IS ASSIGNED TO USER STU3 
 , GROUP STCGROUP
$HASP373 STU3053  STARTED
+(broker53) 0 BIP9141W: The component was started.
+(broker53) 0 BIP2001I: The IBM Integration Bus 
service has started
 process ID 33554919.
+(broker53.default) 0 BIP2201I: Integration Server started: process '67109
442
 196'; thread '0'; additional information 'broker53', '76eb7f2d-e800-00
 00-0080-974c271866d2', 'default', 'true', 'Q4A3', 'false', 'ARG5D651',
  'ARGO53', '********', 'false', 'f9c27f2d-e800-0000-0080-974c271866d2'
 , '/local/argo/driver/drv3', '/local/argo/tgrp53/broker53'.
+(broker53.default) 0 BIP9137I: A work manager has been registered by R
443
 RMS registration services, work manager name is BIP.STU30532.006710919
 6.IBM.UA  .
    2. Display the address spaces. The following example shows the display of an integration node with one integration server: 
      D OMVS,U=STU3
BPXO040I 18.49.59 DISPLAY OMVS 446 
OMVS     000E ACTIVE          OMVS=(68,05) 
USER     JOBNAME  ASID        PID       PPID STATE   START     CT_SECS 
STU3     STU30531 0069   33554696   33554919 HR    18.49.15      2.217 
  LATCHWAITPID=         0 CMD=bipbroker broker53 
STU3     STU30532 03FD   67109196   67109222 HR    18.49.23     19.816 
  LATCHWAITPID=         0 CMD=DataFlowEngine broker53 76eb7f2d-e800-00 
STU3     STU3053  0036   33554768   83886483 HRI   18.49.08       .653 
  LATCHWAITPID=         0 CMD=bipservice Q4A3BRK AUTO 
STU3     STU30532 03FD   67109222   33554696 1W    18.49.23     19.816 
  LATCHWAITPID=         0 CMD=bipimain DataFlowEngine broker53 76eb7f2 
STU3     STU3053  0036   83886483          1 1WI   18.49.08       .653 
  LATCHWAITPID=         0 CMD=/local/argo/driver/drv3/bin/bipimain bip 
STU3     STU30531 0069   33554919   33554768 1W    18.49.15      2.217 
  LATCHWAITPID=         0 CMD=bipimain bipbroker broker53

Results

The infrastructure main program bipimain is the first process in every address space. It starts bipservice, bipbroker, or DataFlowEngine as the second process in the same address space. For each integration server, an additional address space is started. In this example, only one integration server is available.

Abend code 0C1 is issued when you try to start the DataFlowEngine on z/OS

Procedure

  • Scenario: The first two IBM Integration Bus address spaces start successfully, but the third address space (the DataFlowEngine) fails to start. The result is an 0C1 abend.
  • Explanation: The DataFlowEngine address space is generated by the admin agent. If the region size is too small, either because an insufficient region size was specified in the procedure, or because the region size was overridden by the z/OS IEFUSI exit, the DataFlowEngine address space might fail to start, and fails with an 0C1 abend.
  • Solution:
    1. Use the IPCS command on the dump (move the dump from the file system to a traditional MVS™ data set if required): 
      verbx vsmdata,'noglobal,jobname(vcp0brk2)'
      where vcp0brk2 is the name of the failing job.
    2. Find the string 'VSM LOCAL DATA AREA AT ADDRESS '.
      The field ELIM gives the available region size and must be greater than 0C800000. If the field SMFEL is not ffffffff, the IEFUSI exit has changed the allowable region size. This value must also be greater than 0C80000.
    3. If you have an IEFUSI exit, check that the exit does not limit the integration node address spaces.
      For example, a commonly used field is OUCBSUBN. This field can be STC or OMVS for the integration node, and indicates how the address space was started.

Error message BIP2604 with return code MQRC_CONNTAG_IN_USE is issued during the start of a message flow on z/OS

Procedure

  • Scenario: Error message BIP2604 is issued with return code MQRC_CONNTAG_IN_USE during the start of a message flow on z/OS: 
    BIP2604E: Node failed to open WebSphere MQ queue [queue name] 
owned by queue manager [queue manager name]
    This message is output every 30 minutes.
  • Explanation: On z/OS, WebSphere MQ supports serialized access to shared resources, such as shared queues, through the use of a connection tag (serialization token) when an application connects to a queue manager that participates in a queue sharing group.

    In this case, a message flow node fails to connect to the indicated IBM Integration Bus queue manager that is associated with the input queue, because the serialization token that it passed is already in use within the queue sharing group.

    This message is for information only. It indicates that serialization is occurring when two or more message flow input nodes try to connect to a queue manager to get messages from a shared queue.

  • Solution: Check whether another instance of the message flow, or a flow using the same serialization token, is already running. If so, no further action is needed. Otherwise contact your IBM Support Center.

After creating or changing a configurable service, you restart your integration node but your message flow does not start, and message BIP2275 is issued in the system log or Windows Event Viewer

Procedure

  • Scenario: After creating or changing a configurable service, you restart your integration node but your message flow does not start, and message BIP2275 is issued in the system log or Windows Event Viewer, indicating that an error occurred while loading the message flow from the persistent store.
  • Explanation: When you change or create the configurable service, the connection properties are not fully validated at that point; the integration node does not attempt to use them to make a connection. For inbound adapters, the connection is made only when the integration node is restarted. Therefore, the properties that you set on the configurable service might be invalid.
  • Solution: Look at the messages following the BIP2275 message to determine if the message flow failed to start because of invalid connection properties.
    For example, in SAP you would see message BIP3414 with a reason such as:
    Connect to SAP gateway failed 
Connect_PM GWHOST= invalidhost.test.co, GWSERV=sapgw00, ASHOST= invalidhost.test.co, 
  SYSNR=00 
LOCATION CPIC (TCP/IP) on local host 
ERROR partner not reached (host invalidhost.test.co, service 3300) 
TIME Fri Nov 28 15:27:32 2008 
RELEASE 640 
COMPONENT NI (network interface) 
VERSION 37 
RC -10 
MODULE nixxi_r.cpp 
LINE 8728 
DETAIL NiPConnect2 
SYSTEM CALL SiPeekPendConn 
ERRNO 10061 
ERRNO TE'
    followed by a BIP3450 message with an adapter error message such as: 
    Connect to SAP gateway failed 
Connect_PM GWHOST= invalidhost.test.co, GWSERV=sapgw00, ASHOST= invalidhost.test.co,
  SYSNR=00 
LOCATION CPIC (TCP/IP) on local host 
ERROR partner not reached (host invalidhost.test.co, service 3300) 
TIME Fri Nov 28 15:27:32 2008 
RELEASE 640 
COMPONENT NI (network interface) 
VERSION 37 
RC -10 
MODULE nixxi_r.cpp 
LINE 8728 
DETAIL NiPConnect2 
SYSTEM CALL SiPeekPendConn 
ERRNO 10061 
ERRNO TE
    This error was detected by the adapter. The following message describes the diagnostic information that is provided by the adapter:
    Connect to SAP gateway failed 
Connect_PM GWHOST= invalidhost.test.co, GWSERV=sapgw00, ASHOST= invalidhost.test.co,
  SYSNR=00 
LOCATION CPIC (TCP/IP) on local host 
ERROR partner not reached (host invalidhost.test.co, service 3300) 
TIME Fri Nov 28 15:27:32 2008 
RELEASE 640 
COMPONENT NI (network interface) 
VERSION 37 
RC -10 
MODULE nixxi_r.cpp 
LINE 8728 
DETAIL NiPConnect2 
SYSTEM CALL SiPeekPendConn 
ERRNO 10061 
ERRNO TE
    This message suggests that the applicationServerHost and gatewayHost properties are incorrect. When you have determined which properties are incorrect, use the mqsichangeproperties command to correct the properties, or use the mqsideleteconfigurableservice command to revert to the properties that were deployed in the adapter. Restart the integration node.

A device allocation error is issued

Procedure

  • Scenario: A device allocation error is issued.
  • Explanation: A likely cause of this problem is that you do not have the correct permissions set on the component file system for the started task ID.
  • Solution: Check the system log; if the problem is caused by having incorrect permissions set for the started task ID, you often see an RACF® authorization failure message, as shown in the following example. 
    ICH408I USER(TASKID1 ) GROUP(TSOUSER ) NAME(FRED (FRED) 959   
 /argo/MA11BRK/ENVFILE   
 CL(DIRSRCH ) FID(01D7C7E2E3F0F8000F16000000000003)   
 INSUFFICIENT AUTHORITY TO LOOKUP   
 ACCESS INTENT(--X)  ACCESS ALLOWED(OTHER ---) 
IEE132I START COMMAND DEVICE ALLOCATION ERROR 
IEA989I SLIP TRAP ID=X33E MATCHED.  JOBNAME=*UNAVAIL, ASID=00A8. 
D J,BPXAS 
IEE115I 11.13.04 2001.212 ACTIVITY 601
    In this example, the started task ID does not have access to the file system component. The ICH408I message shows:
    • The file that the task is trying to access
    • The user ID that is trying to access the file
    • The permissions that the ID is expecting to have (INTENT in the message)
    • The permissions that the ID actually has (ALLOWED in the message)

    You can use this information to correct the permissions, then reissue, in this example, the request to start the integration node. This type of message is produced if the user who is issuing the command (which might be to start the integration node, or to submit JCL to start one of the utility jobs) does not have the correct file system permissions for the file system component. Use the ICH408I information to rectify the problem.

    Another possible reason for authorization failures is inconsistencies in the RACF definitions for a user ID in the MVS image and the OMVS segment. Also check with your system administrator that the RACF ID that is used on MVS has a corresponding OMVS image created.

Windows fails to recognize IBM Integration Bus digital signatures: "Unknown Publisher"

Procedure

  • Scenario: User A installs IBM Integration Bus, and can run all programs (mqsi*.exe, bip*.exe, including the command console launcher). User B is created and given appropriate privileges. When User B runs an executable file such as the command console launcher, a window opens and reports that the executable file is from an unidentified publisher.
  • Explanation: The operating system has not installed the appropriate digital certificates for User B.
  • Solution: User B must manually install the certificates:
    1. In Windows Explorer, navigate to the bin directory for the IBM Integration Bus installation; for example, on 32-bit systems, C:\Program Files\IBM\MQSI\7.0\bin
    2. Right-click any .exe file to open the Properties window.
    3. Click the Digital Signatures tab.
    4. Select the appropriate certificate from the list, then click Details. The Digital Signature Details window is displayed.
    5. Click View Certificate. The Certificate window is displayed.
    6. Click Install Certificate and complete the steps in the wizard. (Click Next, Next, Finish, then click OK.)
    7. Close the Certificate window. You return to the Digital Signature Details.
    8. Select the countersignature from the list, then click the Details button. A new Digital Signature Details window is displayed. You can repeat the preceding steps to install other certificates.

The create command fails, and error message BIP8022 is issued

Procedure

  • Scenario: Error message BIP8022 is displayed when you use the mqsicreatebroker command on Windows, even if the supplied user name and password are correct.
  • Explanation: The Microsoft component "Shared File and Printer Services" is required.
  • Solution: Correct this error by installing the "Share File and Printer for Microsoft network" service on the Windows system.

Your integration server restarts repeatedly with a JVM Startup failure

Procedure

  • Scenario: When you start the DataFlowEngine it continually starts and stops, displaying errors BIP2116E and BIP7409S in the log: 
    BIP2116E: IBM Integration Bus internal error: diagnostic information 'Fatal Error; exception thrown before initialisation completed', 'JVM Startup'
BIP7409S: The integration node was unable to create a JVM. The return code indicates that an unrecognized option was passed in to it.
  • Explanation: When you start an integration server, it creates a Java virtual machine (JVM) for executing Java user-defined nodes, and its creation failed due to an incorrect JVM option.
  • Solution: Correct the JVM option by completing the following steps:
    1. Stop the integration node.

      mqsistop integrationNodeName

    2. Check the jvmSystemProperty value of the failing integration server.

      mqsireportproperties brokerName -e egName -o ComIbmJVMManager -n jvmSystemProperty -f

    3. If the jvmSystemProperty has an invalid option, correct or reset its value.

      mqsichangeproperties brokerName -e egName -o ComIbmJVMManager -n jvmSystemProperty -v "" -f

    4. Start the integration node.

      mqsistart integrationNodeName