@flaxking said in Which is easier to learn Ansible or Chef or Puppet:

The whole container/CM thing is kind of at a weird place right now. With containers, we've sort of migrated back to the golden image thing, just with the pressure to create an image using a config file. I don't really like images on docker hub that don't have links to the dockerfile.

Yes, images that can't be tweaked without rebuilding. It's has benefits, and a lot of caveats. Having to deploy new containers for every patch is pretty silly. And so many things can't be immutable, including the platform on which the containers run.

At the end of the day, containers are either mutable, or immutable. The former need CM inside the container, the later need it outside the container.

Also, thank you for the followup. You never know when someone else will run into the same thing.

http://blog.takipi.com/deployment-management-tools-chef-vs-puppet-vs-ansible-vs-saltstack-vs-fabric/

@johnhooks said:

@scottalanmiller said:

@johnhooks said:

You don't think $5,000 a year for 100 nodes is reasonable?

Not so much, no. When you consider that the assumption is that many of those nodes are $5/mo to operate. That could be nearly 10% of your entire server budget!

lol plus thats the base price with only 30 days of support and no SLA ....

Exactly. Those prices get a bit crazy. $5/year/server would make a lot more sense. There are many cases where I would pay more for Ansible than I would for the server itself!

Now to make our first organization, we will make a test one to get started.

chef-server-ctl org-create SHORTNAME LONGNAME --association_user USERNAME -f filename

That's the format, here is a real command:

chef-server-ctl org-create test "Chef Testing NonProd" --association_user scott -f scott.pem

Really quite simple since NVM is so simple itself. Just used the standard curl + bash command to run the NVM install of the version of my choice (handling updates will have to come in a later version.) I use the install directory relocation option and that puts NVM into /usr/local/nvm rather than into the user's own folder. Then just source the newly installed nvm.sh file and run the NVM Node.js install command as usual. Voila.

There we go, all set. Hopefully that will help some people get started with Chef since it is basically really simple but lacking a few specific things that you "just have to know" because Opscode does not document them (a gap in the Chef 12 documents) it is very hard for no reason.

It is the enterprise / premium features that are limited to 25 nodes (on the honor system.) The base server is completely open source and unlimited.

@Aaron-Studer said:

@scottalanmiller said:

@Aaron-Studer said:

Should I learn Chef or Puppet?

Good question. The answer is squarely... It depends.

Do you want hosted or on premise? On Prem

How many machines do you want to support? 25ish

Do you need a graphical interface? Yes.

What platforms do you want to support? Windows, and Linux.

Then Puppet as Chef doesn't do graphical.