Current-plan recovery principles
Tivoli Workload Scheduler for z/OS is designed so that in most error situations, the current plan can be automatically recovered without any action required by you.
To achieve the recovery, the following data sets are used:
- EQQCP1DS
- Primary current plan data set
- EQQCP2DS
- Alternate current plan data set
- EQQSCPDS
- Current® plan copy used to produce the Symphony™ file.
- EQQCXDS
- Current plan extension data set
- EQQNCPDS
- New current plan data set
- EQQNCXDS
- New-current plan extension data set
- EQQJTnn
- Current and inactive job-tracking logs
- EQQDLnn
- Current and inactive dual job-tracking logs
- EQQJTARC
- Job-tracking archive log
- EQQCKPT
- Checkpoint data set.
However, the descriptions that follow use these logical terms to describe the CP and its associated data sets:
- Current plan
- Used when describing the current plan in general. The current plan consists of the active current-plan data set and the extension (CX) file.
- Active current plan
- Refers to the current-plan data set that is currently in use within Tivoli Workload Scheduler for z/OS. It is either EQQCP1DS or EQQCP2DS. Every time a current plan backup is performed, Tivoli Workload Scheduler for z/OS switches the active current plan to the other data set. Managing the Workload describes the current plan backup process.
- Current plan extension
- Refers to the file that contains special resource information. The extension file is held in a data space, which is backed by DASD file EQQCXDS. When a current plan backup is performed, the data space is refreshed to the DASD file.
- Inactive current plan
- Refers to the current-plan data set that is not currently in use. It contains a backup copy of the current plan. It is either EQQCP1DS or EQQCP2DS.
- New current plan
- Refers to a new version of the current plan, which is created by one of the daily planning batch jobs. It refers to the EQQNCPDS data set and EQQNCXDS data set.
- Current and inactive job-tracking log
- Refers to the data sets used by IBM Tivoli Workload Scheduler for z/OS to log updates to the current plan and to record audit information for requested files. You must use at least two job-tracking logs, referenced by ddnames EQQJT01 and EQQJT02. You can use up to 99 job-tracking logs. The JT logs are used in a cyclic manner, and IBM Tivoli Workload Scheduler for z/OS automatically switches to the next available JT log after a CP backup. The data from the inactive data set is copied to the archive log, and the data set is emptied in preparation for future use. You should use at least 5 job-tracking logs. This is the default number on the JTLOGS keyword of JTOPTS.
- Current and inactive dual job-tracking log
- If the dual logging function is requested, IBM Tivoli Workload Scheduler for z/OS duplicates the JT records in the corresponding dual JT log. Dual logs are switched at the same time, and in the same sequence, as the JT logs. So the number of dual job-tracking data sets is determined by the number of normal job-tracking data sets.
- Job-tracking archive log
- Represents the accumulated job-tracking data since the new current plan was created. When the JT log is switched, the data from the inactive data set is appended to the archive log. The archive log is copied to the data set referenced by the EQQTROUT ddname during the daily planning process. When IBM Tivoli Workload Scheduler for z/OS takes over the new current plan, the archive data set is emptied.
- Checkpoint
- Refers to the EQQCKPT data set, which contains information about the current status of the IBM Tivoli Workload Scheduler for z/OS system, including which current plan and job-tracking data sets are currently active.
- Symphony file
- Represents a local plan for a set of fault-tolerant workstations and is updated accordingly to the local and current plan changes.
The basic principle of IBM Tivoli Workload Scheduler for z/OS current plan recovery is that if the active current plan becomes unusable for any reason, IBM Tivoli Workload Scheduler for z/OS should always be able to re-create an up-to-date current plan from the backup current plan and the various job-tracking logs. The way IBM Tivoli Workload Scheduler for z/OS actually carries out this task is described in Current-plan recovery processing.