Middleware News

When you configure inetd or xinetd with WebSphere MQ 5.3 for Linux on Intel or zSeries systems, you may find that the amqcrsta TCP/IP responder program will fail with an FDC showing Probe Id XY439010 from Component xcsProgramInit. Cause The FDC will show a Major Errorcode of MQRC_ENVIRONMENT_ERROR, which means that it is not running with the threading model that MQ expects. WebSphere MQ V5.3 uses the LinuxThreads model rather than the newer NPTL threading model available on many newer Linux distributions. Resolving the problem Configure WebSphere MQ to use LinuxThreads by setting the following environment variable wherever the queue manager, its programs, or user applications will run: export LD_ASSUME_KERNEL=2.4.19 Be aware that this variable can cause problems for some applications which require NPTL threading. In particular, refer to the item on RPM database corruption when using this environment variable. For this reason the IBM® MQ programs can use an alternate v...

After you upgrade from WebSphere MQ V6 to V7, many instances of the errors AMQ6090 and AMQ5037 are shown in the queue manager error logs. Symptom The following errors are shown in the WMQ queue manager error logs. AMQ6090: WebSphere MQ was unable to display an error message 6287. EXPLANATION: MQ has attempted to display the message associated with return code hexadecimal '6287'. The return code indicates that there is no message text associated with the message. Associated with the request are inserts 0 : 0 : AMQ5037: The Queue Manager task 'ERROR-LOG' has started. EXPLANATION: The Utility Task Manager, process ID (708760) type(203), has started the ERROR-LOG task. Meanwhile, FDC files are generated which correspond to the AMQ5037 errors: | Probe Id :- XC022001 | Component :- xcsDisplayMessage | Program Name :- amqzmur0 | Major Errorcode :- zrcX_TASK_STARTED | Probe Type :- MSGAMQ5037 | Probe Description :- AMQ5037: The Queue Manager task 'ER...

You install WebSphere MQ for Linux . You try to create a queue manager using the crtmqm command and It fails because it could not open the MQSeries message catalog (amq.cat). Symptom [mqm mqm]$ crtmqm -q F.QMGR AMQ6090 WebSphere MQ was unable to open a message catalog to display an error message for message id hexadecimal %6, with inserts %1, %2, %3, %4, and % 5. Cause Directory /usr/share/locale/C , had insufficient permissions (700). Resolving the problem Change directory /usr/share/locale/C to 555 (r-xr-xr-x). These problems are generally permissions problems. Verify the file permissions on the WebSphere MQ message catalog (amq.cat) and every directory element for the path. By default the WebSphere MQ message catalog will be in: /usr/share/locale/ /LC_MESSAGES/amq.cat Note: Substitute your language for

Your queue manager ended as a result of the server being rebooted. You restarted your queue manager and noticed many FDCs with probe id XC338001, component xehAsySignalHandler: Symptom Example of FDC: Probe Id :- XC338001 Component :- xehAsySignalHandler Major Errorcode :- xecE_W_UNEXPECTED_ASYNC_SIGNAL Comment1 :- SIGTERM +------------------------------------------------------- MQM Function Stack xehAsySignalMonitor xehHandleAsySignal xcsFFST Cause The server was rebooted without first shutting down the queue manager. Resolving the problem This is working as designed. You need to shut down your Queue Managers before rebooting the system to avoid these FDC reports. Additional information Usually there were many reports of this type of FDC from various WebSphere MQ processes and all these reports were created within a few seconds. The server was rebooted without shutting down the queue manager. Under these conditions it is expected, and desirable behavio...

In the reported case, MQRC_ENVIRONMENT_ERROR 2012 is coming from the local queue manager when a WebSphere MQ Java client attempts to do an MQCTL call. Symptom Suddenly the WebSphere Application Server (WSAS) stops getting messages from the MQ queue. The WSAS control region log contains: JMSCMQ0002: The method 'MQCTL' failed. Caused by: com.ibm.mq.MQException: JMSCMQ0001: WebSphere MQ call failed with compcode '2' ('MQCC_FAILED') reason '2012' ('MQRC_ENVIRONMENT_ERROR'). Cause A slip dump showed that the async consume thread TCB ended. In logrec, the TCB hit an ABENDS0D7 RC25 abend in BMC module MQTRFGST followed by an ABENDS0C1 in BMC module MMAHPCST. The S0D7 and S0C1 are the reason for the async consume thread ending, which causes the MQCTL to fail with MQRC 2012. Diagnosing the problem A slip was provided by the support center to get a dump in CSQBSRV when the MQRC 2012 occurs. Resolving the problem Recycling ...

 The memory % in "show load" can be significantly higher than the memory % in "show memory". Why are they different? What do they mean? Symptom The WebSphere DataPower WebGUI status value "System Usage "(CLI command " show load ") has a memory % that can be significantly higher than the status value "Memory Usage " (CLI command " show memory" ) memory %. What accounts for this discrepancy? Cause When you display "Memory Usage " from webGUI, notice the various categories of memory values: Total Memory Used Memory Free Memory Requested Memory Hold Memory The memory value in "Memory Usage" from the webGUI (" show memory " in CLI command) is based on Used Memory / Total Memory. The memory value in System Usage from the webGUI (" show load" in CLI command) is derived by Requested Memory / c...