Validate All on 10.2

[TedCronin]

If I validate all on the batch job window/dialog, i see no configuration issues, but if I validate all on the Batch Job Manager, there are configuration issues. Why the discrepancy?   Should these be consistent with one another, or is this the expected behavior?

Also, when running the same test on a user machine, I see two tables both Domain Checks that are having issues, when on my machine, there are no issues with these tables.

This is 10.2.

[MichelleJohnson]

In regards to the first question, Batch Validate auto-resources the batch job when you open it.  Therefore when you validate all, you may not see any configuration errors.  Batch Job Manager does not automatically resources the batch job you open.  (You would have to do that thru the Workspace button.)  Therefore if your batch job is pointing to a data path that you don't have on your computer, then you would have validation errors.

For the domain check, did you double check that the data being validated was the exact same resource (version)?

[TedCronin]

In regards to the first question, Batch Validate auto-resources the batch job when you open it.  Therefore when you validate all, you may not see any configuration errors.  Batch Job Manager does not automatically resources the batch job you open.  (You would have to do that thru the Workspace button.)  Therefore if your batch job is pointing to a data path that you don't have on your computer, then you would have validation errors.

For the domain check, did you double check that the data being validated was the exact same resource (version)?

Good explanation.  One is implicit and the other is explicit.  I restate some of your explanations below along with Questions/observations...

So, the Batch Validate Implicitly does the association to the version and the server for that matter, the domain check validation issue is only an error when the data is a table, when the Domain check is a layer, there are no issues with the validation.  This is the way it works unless I allow Reviewer to add the data to my MXD, problem is we have Editing Templates, so now I have two instances of these tables, 1 set from the edit template, and 1 set added from Reviewer, and this is the behavior for the Batch Validate for Table domains.  I do not need to allow Reviewer to add all the other layers associated with the checks.  Seems inconsistent.  Needed for Tables, not for Layers.


The Batch Manager does not do the association unless I explicitly set the Workspace and Version, and I need to to this association to the Layer/Table and the actual Reviewer Batch Job Resource.  Why does it have to be done twice, seems like when I change one workspace, all should get changed for both the Layer/Table and Reviewer Batch Job Resource.

Right now our process is to build the checks on a Test Server (GIS2), then reset via the Batch manager to Set the Workspace from GIS2 to GIS3, Save as a New Batch Job.  When the Checkers need to run a Batch Job, they then follow that up with a Change to a Specific version name.  It seems like this process could be streamlined a bit more to follow how the process works inside of the Batch Validate, where I can run a Validate in Batch manager then just go right to a Save As to capture the version/server info as part of the layers/Checks...


On an unrelated note, we are experiencing Session jumps, where we are up to Session 418, and we have not created 417 Test Sessions, not sure what's going on here.

Also, a nice to have...  The ability to Sort by Owner/Version in the Select Version pop up when Change Version used for Batch Manager.  Currently the Version and Owner are combined in one field.