jobs, based upon the date of submission, with certain params

[gglau]

There are only two choices.

1) Implement multiple Control-M jobs, with each one passing the required parameters to backup command based on day of week/month. Only one job will be ordered per day which will not affect task license.

2) Control-M job triggers a script, which has logic to determine day of week/month for passing proper parameters to backup command.

The person who says NO to choice one should come up with the script for choice two.

[markf]

Data Protector's own scheduler is rarely used and a lot of companies use Control-M to deal with the particular problems presented by DP.

The good news is that you can solve these issues, the bad news is that there is some initial work involved.

We run the backups like this -

omnib -datalist your_datalist_name_here -mode full -protect weeks 8

but each backup job will just have -

omnib -datalist %%MEMNAME -mode %%BU_TYPE -protect weeks %%BU_RETEN

(n.b. database backups have modified commands, your DP people will be able to advise).

Build the jobs so -

1. In the File Name field put the name of the datalist that you want to run.
2. On the command line use the command shown above.
3. Saturday through Thursday run the following command on the Control-M Server -

ctmvar -action set -var "%%#\BU_TYPE" -varexpr Incremental1 -quiet

or just incremental (ask the DP people) and then on Friday's -

ctmvar -action set -var "%%#\BU_TYPE" -varexpr Full -quiet

4. Saturday through Thursday run the following command on the Control-M Server -

ctmvar -action set -var "%%#\BU_RETEN" -varexpr 2 -quiet

On Friday's -

ctmvar -action set -var "%%#\BU_RETEN" -varexpr 5 -quiet

Fourth Friday of the month -

ctmvar -action set -var "%%#\BU_RETEN" -varexpr 53 -quiet

For added security you may actually want to make the backups dependent on these 'setup' jobs.

The above will work fine but you may want to consider the following refinements -

1. Create Quantitative Resources equivalent to the number of tape drives available and define 1 per backup job.
2. Create an exclusive Control Resource per backup job (prevents the same backup running at the same time as another).
3. Create exclusive Control Resources for things like firewalls (i.e. if you want to run single stream).
4. The DP return codes are poor and don't really help. I have a script that I run as a POSTCMD that reads the summary card at the end of the DP output and it checks that the "Completed Disk Agents" is equal to the "Disk Agents Total" (if it is not I echo out a statement that I look for via an On/Do check).
5. You may want to consider a couple of reruns - the backups often fail and work fine a hour later.