- Containerise services
- Discovery mechanism
- Surface info, following discovery - json endpoint?
- Routing
- DNS
- Port running on
- Stuff for displaying homepage:
- Service name
- URL
- Icon
- Whether service works offline
- Monitoring?
- Routing
- Create new VM on hive for housing containers
- Start with things which encounter operational problems (eg tfluke app)
- Next focus on legacy stuff on prometheus (eg DNS)
- Then anything not on daedalus
- Finish with remaining daedalus services
- Once completed, could look at removing virtualisation layer and running containers on bare metal