Same Python Packages Installed Across Multiple Dedicated Workers

[mchamps]

For those managing multiple workers in larger scale Alteryx Server deployments how to you ensure that the same Python packages (to be used by the Virtual Environment) are installed globally across all workers?

Can this be done via command line?

• https://help.alteryx.com/developer/2018.1/Python/use/VirtualEnvironment.htm
• https://community.alteryx.com/t5/Alteryx-Designer-Knowledge-Base/How-To-Use-Alteryx-installPackages-...

[raychase]

I'd love to hear some advice on this. It would also be useful if installed packages would persist through the installation of a newer version. In the past, I've had to manually reinstall third-party packages after every upgrade. At the very least, it would be nice to know which folders to backup, so they could be migrated over after the new install.

From my experience, any non-included R libraries persist with an upgrade, so I'm not sure why Python packages do not.

[yuriy]

@mchamps were you able to find the answer or test out some solutions?

@Alteryx_KB - can someone from Alteryx comment on this question?

Thank you

[TimN]

Hi,

We have two worker nodes - Windows Virtual machines as servers.  We looked at cloning the old node to the new (you'd need to change a few parameters) but Alteryx support did not support it.  We ended up creating a requirements file with all the libraries and installed on the new server.  We keep the requirements file up to date for Server Upgrades.