Hi,
in order to streamline inventory management at Brainly we have developed a dynamic inventory plugin:
https://github.com/brainly/inventory_tool
The basic idea is that if you want to automate things like DNS or have >1000 machines, host_vars directory management starts to be tedious. This tool is intended to fill in the gap between big and complex tools like Collins (https://github.com/brainly/inventory_tool) and basic file management. You can find much more details in inventory tool’s README.md
Could you please take a look at it and tell what you think ? I would be very gratefull for feedback, whether it could be useful for others.
Thanks in advance
pr
So this is basically a different inventory topology but also in a YAML file?
I don’t think this is something we’d incorporate in core, but it’s easy to use any inventory source, and it’s always been the thinking that if a particular inventory source doesn’t work for someone, that’s why it’s pluggable, so good deal.
I personally would like to see something like Cobbler’s CLI (sorry, biased a bit!), but just as an inventory data source without the provisioning bits – though a lot of people sync inventory with cloud so it may not always apply or be needed for them.
Hi,
thanks for a reply.
Yes, basically it puts everything into one file + adds IP auto-assignement + stricter syntax checking. You can also much easier interface with the inventory (load->update->save) from within scripts.
I am not 100% happy with the commandline either - each time script runs it has to serialize and de-serialize quite big file, I also think that I have reached argparse’s limits and parts of the syntax seem clunky.
It would take time to write it, and it would bring little benefit to our current setup. Do you think it could increase the chances for including it into the core ?
Thanks
pr