Package Explorer

Deployment Errors

In some cases, for example, when scripts contain incorrect SQL syntax or code, deployment errors can occur. When this happens, Combine will complete the execution to the nearest synchronization point (please see 'Mixed mode' in Execution Modes for details about sync points), and then will behave according to "Promt On Error" option.

If it is set to 'False' then Combine will use "Cancel Execution" option to handle the occurred errors.

If it is set to 'True' then Combine will pause. In other words, Combine will not continue to deploy additional scripts until you instruct it to continue. The dialog below will appear when Combine has reached the next synchronization point.

This dialog contains a list of failed database on the left pane with a corresponding script and error details on the right pane. When you select a database on the left hand side of the dialog the corresponding script name and error messages are showed on the right pane.

Figure 101a:  The deployment error dialog. Using this dialog, the user can select how to proceed with package execution, as well as view all execution errors for each failed database.

Also this dialog allows you to select appropriate action to handle an error for every database. Actions can be selected either by a context menu for selected database or by chosing radio button state on the right pane (radio button state is set according to the selected database). Desired action may be set as for a single database as for all databases on a particular server and even more -- for all failed databases on all servers.

Figure 101b:  The deployment error dialog actions.

Possible actions are:

1. Re-run execution starting on the failed script.
The failed script will be re-executed and then package execution will be continued as usual. This feature may be used in case of recoverable database errors, for example -- no database connection, no access rights, non-existent database object, etc.

2. Skip the failed script.
The failed script will be ignored and then package execution will be continued as usual from the next script. Usually this option is used if user performs manual operations to correct any issues, manually recovers database to a correct state which should be at the end of the failed script execution and then proceeds to deploy the rest of the scripts in the package.

3. Skip all scripts up to the next sync point.
The failed script and all subsequent scripts up to the next sync point will be ignored. And package execution will be continued as usual starting at the first script after the synchronization point.

4. Cancel execution on the selected database.
Combine will just cancel execution on the selected database. No more scripts will be executed against that database.

Combine DOES NOT REVERT any changes which may be done to a database before the script fail. If you restart or skip the failed script -- you will do it at your own risk.

The dialog has two buttons -- 'Continue' and 'Cancel'.
If you press the 'Continue' button Combine will resume package execution and will start to do selected actions for failed scripts (i.e. will re-run/skip scripts or cancel particular databases).
If you press the 'Cancel' button, Combine will stop the package deployment at all and will not deploy any more package scripts on any databases.

© 2001-2018 JNetDirect, Inc. All Rights Reserved.