Continuous Deployment of Forest client

[LesnyRumcajs]

Issue summary I imagine here that we would have a dedicated VPS on AWS that would get upgraded on successful merge to main. We could expose Grafana for everyone to see (if we are proud of the results) and in a way have continuous proof that everything is working fine. On possible incidents, we could send notifications.

Pros:

• Real-time view on Forest state in a real network,
• Shiny dashboards for everyone to see - it may be a tipping point for someone (probably not until we are competitive with features).
• If we would expect someone to really use Forest, it’d be best to prove that it’s continuously working (and doesn’t get e.g. OOMed after 24 hours)

Cons:

• Maintenance time, especially in the early stages.
• Dedicated VPS cost.

Task summary

• [x] Create a dedicated VPS on AWS running Forest on mainnet,
• [x] Implement a workflow that on a successful merge to main, the node on AWS is upgraded (consider using self-hosted GH runners, may facilitate this upgrade),
• [x] Slack notifications on incidents
• [ ] Grafana dashboard

Acceptance Criteria

• [x] On a new merge to main, the Forest node on AWS gets upgraded
• [x] Forest team gets notifications on failures
• [ ] Community can see the status of the node

Other information and links

[lemmih]

Almost done. AWS turned into Digital Ocean. Still need a public dashboard.

[LesnyRumcajs]

We have conitnuous deployment for calibnet, mainnet and butterflynet. Public dashboard will likely not come there (but it may appear somewhere else).