Feat/telegram health monitoring

[Seth-Schmidt]

Issue #13

Checklist

• [x ] My branch is up-to-date with upstream/develop branch.
• [x ] Everything works and tested for Python 3.8.0 and above.
• [x ] I ran pre-commit checks against my changes.
• [x ] I've written tests against my changes and all the current present tests are passing.

Current behaviour

Snapshotter-lite does not have a proper health check endpoint in core-api, and it does not currently have any means of externally reporting issues to node operators.

New expected behaviour

The /health endpoint in snapshotter.core-api requests data from the powerloom-reporting-service (if enabled) to determine the health status of the node, and it will return a failed status code if there are any issues with the node. It checks for the last time the node has pinged the reporting service, as well as if there are any issues reported to the service.

Additionally, the node operator now has the option to provide a telegram chat-id on startup to enable automatic telegram notifications if any issues arise while snapshotting. This feature requires the telegram-reporting-host endpoint to be active and set in the .env.

Change logs

Added

• Telegram notifications in the case of missed snapshots and failed epoch processing

  Changed

• Core-api /health endpoint will monitor the health of the node using data from the reporting service

• Updated snapshotter.utils.callback_helpers to include telegram reporting

Deployment Instructions

Deployment pattern has not changed, but there is now a prompt to enter an optional telegram chat-id on startup.

[xadahiya]

@Seth-Schmidt let's deploy this service on some URL, finish the final testing with latest code, and ship this.

[Seth-Schmidt]

Closing with replacement: #20