Schedule rewards show zero

[bajtos]

I upgraded my Station Desktop instance to v1.7.0 and it started to show 0 scheduled rewards. Downgrading to 1.6.1 did not fix the problem. After downgrading to 1.6.0 I can see my scheduled rewards again.

How to reproduce the problem:

FIL_WALLET_ADDRESS=0xE5975d52558Dc1B4c5249dB5434b585a0Ed69E79 npm start 2>/dev/null

^^^ That does not work, because npm start hard-coded FIL_WALLET_ADDRESS to 0x0...dEaD.

Here is the corrected command:

FIL_WALLET_ADDRESS=0xE5975d52558Dc1B4c5249dB5434b585a0Ed69E79 STATE_ROOT=.state CACHE_ROOT=.cache node ./bin/station.js 2>/dev/null

[bajtos]

I think the problem is caused by the fact that we wait for 10 minutes before we run getContractsWithRetry for the first time after startup.

Do you remember why we implemented it this way? I suspect that we wanted to avoid DDoSing w3name in case Station Core instances frequently restart. Or more precisely, if we frequently restart the main zinnia loop. Now that we moved contract updater outside of the main zinnia loop, I think we can change this behaviour.

A few more alternatives that come to my mind:

• We can hard-code a seed list of contracts into Station Core source code.
• Station Core can persist the list of contract addresses in a state file, so that we have a relative up-to-date list after restart.

[juliangruber]

I think the problem is caused by the fact that we wait for 10 minutes before we run getContractsWithRetry for the first time after startup.

This was so that when Voyager crashes, we don't fetch new contracts. Now that they are decoupled, and we would only fetch contracts when Core restarts, let's move the wait afterwards again. Will PR

[juliangruber]

#477