This feels straight up wrong. KvmVmScore (sounds generic) depending on CloudInit (specific) is a crime against good architecture and naming.
Why do we need that? It is yet another dependency? Can't we pass the cloud init any other way when creating the vm? Do we need that for cloud init as I'm assuming or something else?
Halfway through the review, many small things and a few bigger things to fix. Overall not terrible. But take the time to step back, understand clearly the code review and revisit the entire p-r with the comments in mind and improve it.
Use args vec, more readable. Also true for all other Command::new calls.
feels like ansible should have a purpose built module here too. Don't be lazy and call shell commands. Calling shell commands through ansible is completely useless as they're not truly idempotent in the way that file and package installations are.
ef418f2f96
Merge pull request 'feat: Disable ipv4 address conflict detection score. This is useful when setting up bonds as the wrong mac may get a dhcp offer and then the system will perceive it as a conflict when it sets up the bond correctly' (#263) from feat/disableDadScore into master
6267c2757f
feat: Disable ipv4 address conflict detection score. This is useful when setting up bonds as the wrong mac may get a dhcp offer and then the system will perceive it as a conflict when it sets up the bond correctly
Code review guide — feat/iot-walking-skeleton
One-line summary. Thin end-to-end thread for the IoT platform (ROADMAP/iot_platform/v0_walking_skeleton.md): kubectl apply Deployment in a …