Bison Backup Procedures

Schedule

The CA7 Bison Backup runstream is run daily Monday - Friday on third shift after the normal Bison production jobs have completed. This normally occurs sometime between 2 and 4am and run for about 2 hours. There are USR requirements on the MVSDP230 job reminding you that CICSL and the BISON production runstream must be completed before releasing these runs. The proper schedule id for each day is listed in the table below plus the CA7 command to use to view the entire job sequence for that day's cycle.

Monday	schid 18	FSTRUC,JOB=MVSDP230,SCHID=18
Tuesday	schid 19	FSTRUC,JOB=MVSDP230,SCHID=19
Wednesday	schid 20	FSTRUC,JOB=MVSDP230,SCHID=20
Thursday	schid 21	FSTRUC,JOB=MVSDP230,SCHID=21
Friday	schid 22	FSTRUC,JOB=MVSDP230,SCHID=22

Contact Information

All except one job which is ULAUTHOR (runs on Mondays) is maintained by John Pedlow. The jobnames prefixed with BKP are backup jobs. The jobs prefixed with TMS are daily maintenance runs for CA1 (our tape management system). Jobs prefixed with CA7 are daily CA7 maintenance type runs. The EREP run is the daily hardware analysis program used by our IBM CE. ULAUTHOR is maintained by Mark Ludwig.

Special Instructions

Backup Stream canceled by Mark Ludwig

If for some reason the backups are canceled, let's say for example Mark Ludwig requested that we don't run the backups, it is critical that the rest of the backup stream which includes the tape management and CA7 runs be processed. Look at the schedule for that day using the FSTRUC command given in the above table to see what runs are in the stream. Normally, you can just demand start the TMSCOPY run with the proper schedule id to pick up the necessary runs. If the tape management jobs are not processed daily the tapes will not be scratched on schedule as they should be.

Swap Errors

The FDR (Fast Dump Restore) utility used to backup the files does not handle SWAP messages properly and the normal procedure regarding SWAP messages can not be followed. Different rules also apply whether the backup is a full volume backup or an incremental backup.

Full Volume Backup (BKPBSNF1 and BKPBSNF2)

When you get a swap request on one of these jobnames, cancel the job and restart it once. If the job gets a swap error again, cancel the job, force complete the job in CA7 to continue the runstream and notify John Pedlow when he comes in so he can create special runstreams to complete the job. Also be sure to offline the drive and log a Remedy problem.

Incremental Volume Backup (BKPBSNIO)

When you get a swap request on this job, cancel the job and restart as many times as needed to complete the run. Also be sure to offline the drive and log a Remedy problem.

Created by Judith Thurston

This file was last updated: 25 February 1998 If you have any comments or suggestions please mail to: opnjudy@acsu.buffalo.edu

This page has been accessed:

times since February 10, 1998.