By Robyn Stillwell
Senior Software Developer
IBM Tivoli OMEGAMON XE for IMS on z/OS |
Recent enhancements to IBM Tivoli OMEGAMON XE for IMS on z/OS improve Open Transaction Manager Access (OTMA) monitoring and bring several powerful Tivoli OMEGAMON XE functions to the 3270-based OMEGAMON interface.
Key enhancements include:
- Simplified configuration with automatic discovery of Open Transaction Manager Access (OTMA) groups and members. Discovers OTMA Cross System Coupling Facility (XCF) groups and members for the monitored IMS systems.
- Support for OTMA within the OMEGAMON 3270 interface. Updates to the OMEGAMON 3270 interface include a new menu option and commands to display OTMA status, and TMEMBER and TPIPE status and statistics. (Complete enhancement details are available in the TechNote: OMEGAMON IMS V560 OTMA support.
- New product-provided situations and exceptions for the Tivoli OMEGAMON XE and 3270 interfaces to monitor OTMA. Situation and exceptions alert you to potential OTMA problems that can hinder application performance and throughput. They can help you monitor items such as transaction instance block (TIB) usage to avoid a TMEMBER flood condition; TPIPE status values to prevent application bottlenecks; and TPIPE queue counts and rates to monitor application throughput.
- Support for super member and flood control improvements provided by OTMA. Super member support was added, as well as TIB usage information, which can help you detect the approach or occurrence of a flood condition.
- Exploit new OTMA information provided in IMS Version 10. Includes TMEMBER time-out value for OTMA send-then-commit responses, TPIPE input message count, and TMEMBER and TPIPE status of stop input.
- Improved data collection. Algorithm enhancements significantly reduce OTMA data collection overhead.
A closer look
Tivoli OMEGAMON XE for IMS provides OTMA information for each monitored IMS system. The IMS OTMA Status workspace (Figure 1) can help you determine the health of OTMA. It summarizes the total number of existing TMEMBERs and TPIPEs, and provides overall queue counts and rates. TIB count and rate have been added to monitor TIB usage and growth rate for all OTMA TMEMBERs used by the IMS system.
Figure 1: The IMS OTMA workspace can help you monitor the health of OTMA.
The IMS OTMA workspace allows you to navigate to the TMEMBER workspace to view status and summary information for each TMEMBER. Items added to the TMEMBER workspace (Figure 2) include the TIB count, TIB rate, TIB used percent, maximum TIB count, super member name and DRU exit name. The time-out value for OTMA send-then-commit message responses has been added for use with IMS V10 or later.
The TIB count, rate, used percent and maximum count attributes let you monitor TIB usage by TMEMBER to prevent an OTMA flood condition, which occurs when the number of TIBs in use reaches 100 percent. This event can lead to serious application delays as TMEMBER input is rejected until the flood condition is relieved. Product-provided situations that use these TIB attributes can create alerts to avoid the flood condition.
Figure 2: The IMS OTMA TMEMBER workspace displays status and summary information for each TMEMBER. Status information includes XCF, user, security and trace status. Summary information includes TPIPE count and queue counts and rates.
A link within the TMEMBER workspace allows you to display TPIPE information for a selected TMEMBER. TPIPE information includes individual TPIPE status values and separate queue counts for asynchronous and synchronous output. The TPIPE input count and TPIPE status value of stop input is available for use with IMS V10 or higher. (See Figure 3.)
Individual TPIPE status values are excellent candidates to use when creating situations, because they can trigger an event informing you when one or more TPIPEs have a particular status value. Product-provided situations are available for some of these TPIPE status values.
Figure 3: The IMS OTMA TPIPE workspace shows individual status values and summary information for each TPIPE. In this example, highlighted TPIPE status values warn that TPIPE P0000001 is stopped, and TPIPE TEST0003 is waiting on an ACK/NAK (status WAITA).
Updates to the OMEGAMON 3270 interface
The OMEGAMON 3270 interface for IMS now displays OTMA information previously only available in the Tivoli Enterprise Portal, the graphical interface used by Tivoli OMEGAMON XE for IMS. A new option in the main menu and real-time commands allow you to display OTMA, TMEMBER (Figure 4) and TPIPE (Figure 5) status and summary information.
Figure 4: The OMEGAMON 3270 interface displays OTMA and TMEMBER status and summary information. This figure shows the first page of TMEMBER output.
Figure 5: The OMEGAMON 3270 interface TPIPE display shows TPIPE information for a selected TMEMBER. The TPIPES for a TMEMBER can be displayed by placing the cursor on the desired TMEMBER and pressing the zoom key (PF11). This figure shows the first page of TPIPE output.
New situations and exceptions
New situations in the Tivoli OMEGAMON XE interface and exceptions in the 3270-based interface can alert you to potential OTMA problems. For example, two product-provided situations can monitor the use of transaction instance blocks (TIBs) to help prevent an OTMA flood condition:
- The first situation (Figure 6) produces a warning event when the percentage of TIBs in use for a TMEMBER is between 70 and 80 percent.
- The second situation (Figure 8) generates a critical event when the percentage of TIBs used reaches 80 percent or higher.
The percentage values used in these situations can be easily modified to your specific needs.
Figure 6: A warning event generated by the first situation lights up the navigation tree. The TIB used percent value is highlighted. In addition, a link is provided to the Events workspace (shown in Figure 7).
Figure 7: The Events workspace reveals the values that generated the event, current values, desired action if any, and related advice.
Advice provided by the Tivoli OMEGAMON situation indicates that TIB growth occurs for many reasons, including a stopped program, unavailable message region, or when message arrival rate exceeds processing rate.
Existing OMEGAMON XE workspaces can help you diagnose the cause of TIB growth:
- The IMS Programming Scheduling Blocks workspace can identify stopped programs.
- The IMS Dependent Regions workspace displays all message regions and their current status to help you determine if a region is available.
- The message enqueue and dequeue rates displayed in the IMS System Information workspace can help you determine if the message arrival rate exceeds the processing rate.
If the warning event for the percentage of TIBs used isn’t resolved within a reasonable time frame and TIB usage climbs to 80 percent, the second situation generates a critical event highlighted in red on the navigation tree (Figure 8).
Figure 8: A critical event generated by the second situation lights up the navigation tree when the TIB usage reaches 80 percent or higher. The TIB used percent value is highlighted.
Another product-provided situation generates a critical event when a flood condition occurs for one or more TMEMBERs (Figure 9). This event can be prevented by monitoring TIB usage using the two situations described above and by taking corrective action to resolve the event.
Figure 9: A critical event generated when detecting a flood condition for TMEMBER HWS is shown in red.
Additional situations and exceptions included with the product update can reveal other OTMA conditions, such as when:
- One or more TPIPEs are waiting on an ACK/NAK for a commit-then-send output response. If this situation/exception persists for a TPIPE, users can investigate the OTMA client to determine why an ACK or NAK response is not occurring or is taking too long.
- The status of one or more TPIPEs has stopped, or input has stopped. Receiving notification of this condition and taking corrective action can reduce application bottlenecks.
- One or more TPIPEs have an output queue count or rate that reaches a specified threshold. This condition can inform you of potential problems with asynchronous output queuing.
- One or more TPIPEs have an input queue count or rate (available in IMS V10) that reaches a specific threshold. This alert can help you monitor input message throughput.
Summary
These updates to IBM Tivoli OMEGAMON XE for IMS on z/OS are available with APAR OA22880 for the Tivoli OMEGAMON XE for IMS component (PTF UA37405 for V410 and PTF UA37404 for V310) and APAR OA23838 for the OMEGAMON 3270 component (PTF UA39158 for V560).
Using IBM Tivoli OMEGAMON XE for IMS on z/OS situations and exceptions to monitor OTMA can help prevent and identify OTMA conditions that can impede application performance. OTMA status and TMEMBER and TPIPE displays can provide valuable information to help you determine the health of OTMA and assist in diagnosing problems.
|
