Using this monitor, you can monitor the status of background processing in your SAP system both for the entire system and itemized by application servers on which background work processes are configured. For a general introduction to background processing, see Background Processing: Concepts and Functions.
In addition to monitoring background processing in general, you can also monitor selected jobs using the Monitoring Architecture (see Monitoring Jobs with the Alert Monitor).
Features
The monitor contains the following monitoring tree elements (MTEs):
MTE Name | Meaning |
SystemWideQueueLength | Number of jobs that are ready to be executed, have start authorization, and have no target server specified for which there are no free background work processes, averaged over all application servers with background work processes. |
SystemWideClassABPWP | System-wide number of background work processes that are reserved for high priority jobs (Class A Jobs) |
SystemWideFreeBPWP | Number of free background work processes in the entire system |
SystemWideTotalBPWP | Number of background work processes in the entire system |
R3Syslog\Background | Messages in the system log for the background processing category; you can set the category in which a message is reported, the message text, and the severity and criticality of the alert using the message ID in transaction SE92 |
Utilization | Percentage of the background processing capacity currently utilized; the value is averaged over the background work processes and, by default, averaged over the last hour |
NumberOfWpBTC | Number of background work processes on an application server |
ErrorsInWpBTC | Number of errors in background work processes since the monitoring segment was created (that is, since the application server was started) |
ErrorFreqWpBTC | Number of errors in background work processes per minute |
EndedWpBTC | Number of background work processes terminated after an error; you can use the process overview (transaction SM50) to determine whether a work process should be restarted after an error |
ProgramErrors | Program errors when executing background jobs |
ServerSpecificQueueLength | Number of released jobs that are explicitly to be executed on this application server, but for which there are no free background work processes |
AbortedJobs | Individual aborted jobs on an application server; a separate red alert is generated for each of these jobs |
Activities
To start the monitor, follow the procedure below:
1. Start the Alert Monitor using transaction RZ20 or choose CCMS ® Control/Monitoring ® Alert Monitor.
2. On the CCMS Monitor Sets screen, expand the SAP CCMS Monitor Templates set.
3. Start the Background Processing monitor from the list by double-clicking it.
Procedure if an Alert Is Triggered
Utilization and SystemWideQueueLength show if there is a serious bottleneck in the capacity of the background processing. The optimal situation is that there is a high utilization of the work processes and a short wait queue length.
Background jobs are usually defined without a target server (Execution target in the Define Job initial screen of transaction SM36). This ensures an optimal distribution of the job workload. Jobs defined in this way therefore do not appear in the ServerSpecificQueueLength, but only in SystemWideQueueLength, meaning that the latter attribute is best for showing capacity problems in background processing.
If a bottleneck exists, it is often useful to divide the existing jobs between more servers. You can do this, for example, by switching dialog work processes to background work processes during the operation mode switch to night operation. Before you increase the number of background work processes, note that it does not make sense to have more than three background work processes for each CPU on a background server, as these work processes will already be fully utilizing the CPU. Set the number of these work processes using the system parameter rdisp/wp_no_btc.
For more information about this topic, see SAP Note 39412 (How many work processes to configure).
An alert for ServerSpecificQueueLength when there is a short SystemWideQueueLengthindicates that the distribution of the jobs is not optimal. You should only specify a target server when it is absolutely necessary that the job is started on this server.
No comments:
Post a Comment