Ways to monitor the COB in T24

Having the necessary tools and knowledge is extremely important for effective COB monitoring. This post shares details about the components we can use to carry out this task.

Monitoring the COB with enquiries

This topic describes the three enquiries you can use to monitor the progress of the COB and the processed records

Enquiry	Description
COB.PROGRESS	This enquiry provides a list of active companies and shows, through a progress bar, how much time remains before completion.
JOB.PROGRESS	This enquiry lists all active and completed jobs by descending start times - the current jobs are at the top. For each job, it displays: The start time and end time. Elapsed time. Contracts processed. Total contracts to process. Throughput (contracts per minute). Individual server rate. The individual server rate is the number of contracts processed in one minute by a server process. The system compares this rate historically to find out if the job slows down as time passes.
JOB.HISTORY	This enquiry allows you to drill down from JOB.PROGRESS to a graph that shows: The elapsed time of the last ten runs of the job. The individual server rate. If the individual server rate figure rises over time, this indicates a problem (often badly sized files).

Monitoring with tools

RT.ANALYSE and REPORT.TIMES

You can use the RT.ANALYSE utility to extract statistics from the table REPORT.TIMES. This table contains statistics from every report that is generated during the COB and any online service.

REPORT.TIMES includes the following statistics:

• Name of the report.

• The BATCH and JOB that generated the reports.

• Start time and end time.

• Elapsed time in seconds.

• Temenos Transact date for which the report was generated.

  
  

Temenos Transact allows external systems to monitor the progress of COB and the results can be viewed in T24 log, as a Data Event Streaming (DES) event, and in the Grafana dashboard.

Monitoring COB using Write Cache

The Write Cache facility is designed to keep transaction level information in the cache and reduce the number of repeated WRITE(s) in a single transaction.

By default, the writes and reads from the disk are cached for all COB and online jobs. The advantages of using the Write Cache include:

• Reducing the number of repeated WRITE(s).

• Reducing the number of redundant reads for database activity.

• Handling bulk transactions.

• A final commit to the disk, ensuring data integrity.

  
  

Switching off Write Cache and Exclusions

Procedure:

1. Go to the PGM.FILE

2. By default all COB jobs are cache enabled. To disable the cache for a COB job, set the job to .NUC.

   
   

READ.WRITE.CACHE

The field READ.WRITE.CACHE in JOB.TIMES is set to Y when the RECORD routine uses CACHE. If there are no records selected, then the READ.WRITE.CACHE flag is set to N. This flag helps you identify whether a specific job has used the cache or not.

COB Status Information

TSA.STATUS provides detailed status information for both the COB and online agents. The status information is updated periodically (at every service heartbeat).

TSA.STATUS now provides an easier, 'one stop' information view through a ready enquiry. It also includes critical monitoring information, such as the number of Read/Writes. This provides you with a more detailed picture of the processing carried out by the agent under review.

COB monitoring allows us to detect patterns and issues related to the services being executed. For banks, it is a priority to ensure a stable process with a minimal error rate during COB in order to serve their customers on time at the start of the day.

If you need help identifying and resolving errors or performance issues during COB, contact our Performetriks team. We’ll be happy to support your daily monitoring.

Keep up the great work! Happy Performance Engineering! #COB #Monitoring #Temenos