Gregory's Server@radwebhosting @incognet Yep, shell script can also work. The nice thing about cloud-init is that if implemented you get AWS, DO and many other providers "for free".
Ok, thank you for the info!
|
7 comments
 @radwebhosting AFAIK AWS, GCP, DO and many other providers use cloud-init for setting up fresh VMs. git clone https://git.savannah.gnu.org/git/guix.git virtual size: 1.82 GiB (1949847552 bytes) @radwebhosting Probably the official guix qcow image is built with --image-size=30G option or something like that. @abcdw Crazy! It’s the qcow2 image hosted on their public sftp mirrors…you’d guess they’d want to minimize it, but I won’t pretend to be any kind of Guix expert! Would you like to access a test account on our system to see how it could be managed for successive deployments/updates/etc? @radwebhosting The real size is still < 1.5G, it's just a virtual size, which is huge. I'm quite limited on time for the next couple weeks, but if you interested to provide Guix OS template for your servers, I would be glad to cooperate on this. (To be clear: I can deploy Guix on servers of almost any provider, I'm interested in first-class support, so other people from the community can get it spinned up in seconds with zero additional steps). @abcdw We’d be happy to support you in setting up the optimal service provisioning interface to best manage your client onboarding projects. We have provided similar custom services to other clients and while we don’t provide support directly to your end-users, we’ll help diagnose and resolve any issues you report, originating anywhere within the supply chain. We’ll be available to help you whether you need it now, in three weeks, or in 6 months…just reference our thread 😃✅ |
@abcdw I'm not sure about free AWS or DO, but I just created a deployable Guix template (bare-bones; from the default qcow2) and noticed its over 30GB uncompressed, which is probably why many providers wouldn't support deployable images.
If you have a configuration you'd like to share or a set of commands you perform to achieve your custom configuration, please share them and I'll try to build it, then package it for deployment to ensure automatic setup is not flawed.