Recovering from insufficient disk space for offload processing

If offload processing terminates unexpectedly when Db2 is offloading the active log data sets to disk, you can recover from this situation.

Symptoms

Prior to the failure, z/OS® issues abend message IEC030I, IEC031I, or IEC032I. Offload processing terminates unexpectedly. Db2 issues the following message:

DSNJ128I - LOG OFFLOAD TASK FAILED FOR ACTIVE LOG nnnnn

Additional z/OS abend messages might accompany message DSNJ128I.

Causes

The following situations can cause problems with insufficient disk space during Db2 offload processing:

  • The size of the archive log data set is too small to contain the data from the active log data sets during offload processing. All secondary space allocations have been used.
  • All available space on the disk volumes to which the archive data set is being written has been exhausted.
  • The primary space allocation for the archive log data set (as specified in the load module for subsystem parameters) is too large to allocate to any available online disk device.
Start of change

Environment

The archive data sets that are allocated to the offload task in which the error occurred are deallocated and deleted. Another attempt to offload the RBA range of the active log data sets is made the next time offload is invoked.End of change

Resolving the problem

System programmer response: The actions that you take depend on what caused Db2 message DSNJ128I to be issued:
  • If z/OS abend message IEC030I precedes Db2 message DSNJ128I, increase the primary or secondary allocations (or both) for the archive log data set in DSNZPxxx. Another option is to reduce the size of the active log data set. Modifications to DSNZPxxx require that you stop and start Db2 for the changes to take effect. If the data that is to be offloaded is particularly large, you can mount another online storage volume or make one available to Db2.
  • If z/OS abend message IEC032I precedes message DSNJ128I, make space available on the disk volumes, or make another online storage volume available for Db2. After you make additional space available, issue the Db2 command ARCHIVE LOG CANCEL OFFLOAD. Db2 then retries the offload.
  • If z/OS abend message IEC032I precedes Db2 message DSNJ128I, make space available on the disk volumes, or make available another online storage volume for Db2. If this approach is not possible, adjust the value of PRIQTY in the DSNZPxxx module to reduce the primary allocation. If the primary allocation is reduced, you might need to increase the size of the secondary space allocation to avoid future abends.