Hi @ollyg, I gave this a quick spin and it works great for me! I actually made a quick prototype with Ansible + TextFSM for a CLI based macsuck: https://github.com/rc9000/ntcsuck.

The one thing that puzzles me: it seems to only work once, then I have to restart the netdisco-web process (I was testing with the classic DBIC_TRACE=1 ~/perl5/bin/localenv plackup -R share,lib,~/environments -p 5005 bin/netdisco-web-fg). I first thought it was because of macsuck_min_age, but I removed the setting and it still only seems to work once. Do two consecutive curls for you write to the database both times? Might also be something on my end, when I have some more time I'll see if I can figure it out.

Thanks @rc9000 !
I didn't see the issue you report myself, in testing (using same plackup command) but will have a poke and see what could be going on.

@rc9000 you could also try ?enqueue=true and see if it makes a difference (I wonder if running the backend job inline in the web process is breaking something - it was never intended to work that way!)

...which reminds me, that when using enqueue=true the implementation sets timestamp to when the job is run, not when you submit the data (which would be better), so I'll patch for that.

Oh nevermind it just doesn't work in netdisco-web-fg, the requests > 1 fail with job log:

      'error': 'failed to report from any worker!'

Probably because of the single process environment. When launching with netdisco-web start it works fine.

@rc9000 but that is strange because that's how I'm developing and testing ... in the foreground. I will have a closer look...

You can also use the command line, btw:
netdisco-do macsuck -d x.x.x.x -p /path/to/file/with/the/data.json

@rc9000 aha! I can reproduce, thank you

pushed updates to the PR which remove the enqueue option for web submit, and also fix a couple bugs