Snowplow initialization fails indefinitely with no way of retrying
Problem to solve
In the event that Snowplow fails to initialize, the existing Snowplow initialization worker continues to try indefinitely and doesn't allow the user to attempt to try again. Which is unlike the Jitsu initialization worker which resets Redis after 3 retries.
The fact that it fails and doesn't end means that it will retry indefinitely and the only way to stop it is to have access to Redis to flush the workers initializing key. It also doesn't help that we don't provide the user with any kind of explanation on what is going on.
Proposal
Updated the Snowplow initialization worker to reset Redis after 3 retries, so the user can always try again and not end up in an infinite loop.
A potential follow-up: Store why it failed in Redis for the frontend to pick up and render to the user. This could be cleared when the user initializes again (backend clears it) or closes the alert (frontend hides the alert using browser sessions).