Hi there
Just watched through the videos on the blog post - pretty excited about this release.
Regarding Tenants - I think I’m gonna have to play with it properly to completely understand the impact. We want to see if this functionality will work for test environments amongst teams - not sure how well this will work at this point. The only real feedback I have for the first video is that the tag id’s appear to be guids, rather than formatted like id’s for the rest of the system (ie “Tags-1”).
Regarding elastic environments - this is super cool. A few points I could think of:
- there is a list of machines on the machine policy page - will this work okay when there is 1000+ machines?
- what happens if a machine comes back online after being deleted for being offline too long? Will it reconnect?
- is the machine policy too detailed to show on the machine details page? (Not sure about this one - it may be just right. First thought was too much detail, and it should be hidden behind a link, but then a bit later I thought it might be right - you dont come to this page that often, and the information there is pretty useful at that stage)
- I think there might be some potential confusion about having different spots to say how your release should work. ie, in the process, you say “deploy to a, b & c”, and thats what you’d expect to happen, but then the machine policy is changed by someone else and all of a sudden its okay not to have it deployed onto machine c, and your deployment in octopus succeeds, but your application dont work. (I hope this makes sense.)
- Deployment page - need to be clearer when a machine is offline rather than just red. Might be worth surfacing some details about expected behaviour before the release starts “Machine is offline but the machine policy says its okay to ignore”
cheers,
Matt
PS - the beta-testing-feedback forum title still mentions 3.3 
Hi Matt,
Thanks for your feedback! We really appreciate it
Regarding tenant tags, we’re investigating potential solutions for this one in terms of UI. We’ll likely continue to use GUIDs on the backend for various reasons, but we don’t want those showing in the UI, so we’ll come up with something more user-friendly.
Regarding elastic environments, the list of machines in the policy page’s sidebar has paging support (10 per page), so that shouldn’t get too out of control. But we’ll script 1000+ machines and make sure the UI scales well before release.
Regarding the cleanup behaviour of offline machines, deleted machines are hard-deleted from the Octopus Server’s database, so they will not reconnect if they come back online at a later point. You would need to re-add them as valid deployment targets. The responsibility is really on the user here, and we’ll be adding more warnings around this behaviour and be more explicit in the UI so this is clear.
The machine policy details/summary was an interesting one and we played around with a few options. We decided to list a brief summary directly under the machine policy dropdown so users could quickly see the policy behaviour that was being applied. For customers with multiple policies, we felt this was important so users could quickly flick through their different policies and see a summary in one click.
Regarding the deployment page and the risk of potential confusion, we completely agree. Your suggestion about showing some details before the release starts will be something we’ll aim to improve in an up-coming alpha, because it’s definitely not obvious enough at present.
Thanks again for taking the time to try out an alpha build - the team is very grateful and we look forward to evolving these features around your feedback.
Cheers
Mark
Just letting you know that we’ve got a public Github issue where you can follow the progress of elastic environments, including mockups showing the direction we’re heading for the next release
Thanks again for the awesome feedback!
Mark