Implement health monitoring

[TimDaub]

Scope

• It's really important for the challenge-services to be running reliably as otherwise invalid operations could pass on the LeapDAO Plasma chain

Deliverables

• Create a service that monitors the health of the event-scanner service by e.g. periodically checking:
  • if event-scanner returns a valid response through a HTTP API
  • if event-scanner is at the latest height of the configured Ethereum chain
  • if event-scanner has stored some new events in its queue
  • etc.
• In cases where the health of event-scanner can be considered damaged, the health monitor should send a message to its operators (e.g. through Slack) to notify them about the issue

Notes

• In the future, it might make sense to implement a restart or self-healing mechanism for event-scanner. Given that various things could go wrong, however, I currently believe this would be premature and difficult to do reliably. So human intervention at this point is more reliable.

Gain for the Project

• Improved uptime for event-scanner

Roles

bounty gardener: @TimDaub / 10% bounty worker: name / 75% bounty reviewer: name / 15%

[troggy]

can these be built in the challenge service itself? Otherwise these checks can be made as a an extra checks in https://github.com/leapdao/leap-peeper

[johannbarbie]

no feedback to question