Hi everyone! We're currently deploying a lot of virtual machines across multiple environments, and while Ansible has worked well for us, we've encountered situations where it's not a great fit—like when our machines aren't accessible during deployment or the setup is just a bit messy. So, we're exploring alternatives that can reside on the VM and pull configurations themselves. I'm particularly looking at SaltStack and Puppet. We don't want to commit fully to config management yet; our main goal is to trigger some Microsoft DSC actions once a VM is up and running, such as installing software during deployment.
I've done some basic work with Puppet as a consumer but haven't tried running or setting up the backend. I'm curious to hear from anyone who's used Salt or Puppet in a similar way, especially regarding the pull model—having the agent come back to check in is a big plus for us. Also, considering that SaltStack is open source but backed by Broadcom, is it wise to consider them given their reputation?
5 Answers
If speed and compliance are important, Salt is super quick and works well in certain environments like FedRamp. Meanwhile, solutions like Chef and Puppet can often feel overhead-heavy and complex.
Have you looked into using ansible-pull? It might save you from completely switching tech. Just a heads up though, I found it doesn't support Windows that well, which could be an issue since you deploy a lot of those.
If you want good Windows and DSC support, I’d recommend Puppet. It’s stable and has a solid community backing. SaltStack is flexible, but with Broadcom's involvement, it feels risky. Also, consider checking out Chef Infra or Rudder for pull-based solutions.
I totally get your concerns about Salt being under Broadcom. That said, it's great for managing large fleets and it’s a pull-based system, which can dodge connectivity issues. Still, it has been stagnant lately in terms of development.
Salt is technically owned by VMware, which is now under Broadcom management. That raises some concerns about its future. Ansible-pull might be worth a shot, depending on your use case. What’s preventing your machines from being reachable during deployments? Maybe a solution like AWX can help.
Related Questions
Can't Load PhpMyadmin On After Server Update
Redirect www to non-www in Apache Conf
How To Check If Your SSL Cert Is SHA 1
Windows TrackPad Gestures