Best Practice maintaining workflows accross DEV, UAT & PRD Servers

Question

Hi all,

I was hoping to hear of other's experience maintaining continuity between the workflows saved across 3 environments (DEV, UAT, PRD). And also best practices on backing up the three environments and workflows within.

If all workflows are saved in DEV. Is there a quick and simple way to migrate them over to other environments without manually downloading all workflows and then saving them up individually?

If changes are made to a workflow in DEV by downloading the workflow, editing and then re-publishing will all of the naming and version control stay the same if you simply save to all three environments once edits are done?

Thanks in advance,

Ciaran

CiaranA · Answer

Hi Mark,

Thanks for your response, and yes that is exactly what i'm referring to.

How might this work in practice, once a team has completed a workflow and want to promote it to the three Servers sequentially?

Thanks,

Ciaran

MarqueeCrew · Answer

@CiaranA,

https://help.alteryx.com/2018.1/index.htm#Constants.htm?Highlight=user%20constants

What I hear you asking about is an enterprise requirement to promote code from one environment to the next and in so-doing, keep the code constant through the move process.  Back in the day, we would lock code and need a 3rd party to move the code from work to build, from build to QA, from QA to STAG and finally from STAG to PROD.  As a developer, you could only move code so far before you were locked out of the move environment.  As code moved from one environment to the next, impact assessments were made along the way to assure that the move didn't negatively impact other systems and that the changes were well orchestrated.  Enter servers and exit the mainframe.

If the difference between the environments is minimal, you can create user constants (e.g. DEV_PATH) to account for the variability in dependencies and you can change a constant to address which environment that the code is acting in.  By switching from one environment to the next in a single location across all of the work, you minimize the code disruption.

Cheers,

Mark