Using the Product in a DB2 Environment › Background DB2 Catalog Extraction › Synchronous Data Gatherer › Harvester Error Processing Control

Harvester Error Processing Control

In the remote possibility that the Synchronous Data Gatherer encounters an un-recoverable problem when harvesting data, diagnostic information will be written to the operating system logrec and a message will be displayed on the server JES log.

Since some problems are repetitive in nature, the Harvester data collector has the ability to limit the number of messages displayed, as well as shutting itself down after a designated number of errors. As with other parameters, these commands are located in the TUNSSP00 member

DB2HVMXM

Abend messages.

Designate the number of abend messages that will be displayed in the server JES log if the Harvester encounters an error.

To limit the number of messages displayed in the JES log and LOGREC, use DB2HVMXM to stop the display of Harvester abend messages.

When the message count exceeds the value specified for DB2HVMXM, the following message displays:

TN3441W Max Harvester ABEND message count exceeded for pppppppp jjjjjjjj

where

• pppppppp is the profile name.
• jjjjjjjj is the job name being monitored.

The default is DB2HVMXM=100.

DB2HVMXA

Harvester abends.

Designate the number of Harvester abends that will be allowed before shutting down the collection of data.

When an error occurs, the Harvester writes diagnostic information to LOGREC and displays the following message on the server JES log:

TN3438W Harvester Abend cccccccc aaaaaaaa dddddddd tttttttt ssssssss pppppppp jjjjjjjj

TN3439I Harvester processing continues. pppppppp jjjjjjj

where

• cccccccc is the abend code.
• aaaaaaaa is the next instruction address.
• dddddddd is displacement of instruction.
• tttttttt is the count of abends.
• ssssssss is the intercept sequence number.
• pppppppp is the profile name.
• jjjjjjjj is the job name being monitored.

Harvesting processing continues until the maximum abend count specified by DB2HVMXA is exceeded. At this time, the following message displays:

TN3440W Max Harvester ABEND count has been exceeded. Harvester processing ended for pppppppp jjjjjjjj

where

• pppppppp is the profile name.
• jjjjjjjj is the job name being monitored.

When the value specified for ABEND count is reached, the Harvester will stop collecting data. If DB2HVMXA is zero, no limit is recognized.

The default is DB2HVMXA=1000.