Mup.js deployment

Hello, we are deploying the application we wrote with mup.js and I think we have some time losses during deployment. I wondered if you have any solutions or alternatives for these. I have itemized the problems I am experiencing below.

  1. mup.js first takes the build locally and then uploads it to the machine or register to install it. Then another bundle process is performed here. I think this process can be combined. When I was using dokku, this process was done directly on the master machine with git push dokku master.

  2. The application we wrote has multiple installations. We also have parallel installations where we separate job machines. For this reason, settings.json and mup.js config files may differ. Can we use different settings with the same build once we get a build specific to Mup.

Hey @receptim, I’ve noticed the same issue with mup.js building locally and then uploading, which can slow things down compared to server-side builds like Dokku’s git push. One way to speed this up is to build the bundle once and reuse it across multiple servers, avoiding repeated builds. For different settings.json files, it’s possible to deploy the same build but load unique settings on each server separately, maybe with a post-deploy script or mounting different config files. This approach keeps the build consistent while allowing custom configs per installation. It’d be great to hear if anyone else has found a more streamlined method for multi-install setups.

1 Like