Feature Request: Condition Check Not Met Results In Success
Currently a condition check not met results in a failure of that deployment. It would be great if there was a package option that makes the condition check not met result in success.
I feel this would be a good addition because I use condition checks to exclude machines that already have a particular package installed. With the current configuration, this results in a failed deployment which skews success/failure ratios. In addition, if the package is being run on a schedule and the schedule is configured to retry failed deployments, it will continue to retry and fail until the retry limit is met.
If I am doing something incorrectly or there is a sufficient workaround for this, please let me know!
1
Comments
Yes you missed a thing, the individual "Error Mode" option for each deploy step.
Default is "Stop Deployment with Error" for each new deployment
If you choose "Continue" the next step kicks in even if the currrent step fails.
Be sure you set the right Error Mode for each step
Unfortunately, the error mode would effect how the step responds in an actual error. If I have a step that runs a script and the script produces an error, I don't necessarily want the deployment to continue or complete with success.
What I am looking for is an option at the package level that stops the deployment with success if the check condition is not met; separate from error handling.
Bump. This would be really useful for us as well. If you have an architecture condition set at the package level (not the step level), it's not like that's going to change for a particular target so why keep deploying to it? I can think of situations where I'd want "no conditions met" to continue trying to deploy and others where I would want it to just stop. So it would be great if this could be set at the package level.
I agree. There should be an option where a Condition Not Met is a success.
I have a very simple package step that deletes a file on the desktop. The result is that I don't want that file on the desktop and it doesn't matter to me if my package step deletes it or if the file never existed.
The "Continue" Error Mode in step Options does not achieve the same result.
There are probably a couple thorough ways to accomplish this, but it's not a critical package step. I threw it together in less than a minute and don't think it warrants more attention.
Thanks guys
Bumping this again. This would be a great feature. I currently have to set a generic command line step to allow it to run successfully.