`TPROXY_ENABLE_VARDIFF=false` makes miners appear as offline

[pavlenex]

Not yet sure about this one but it seems if ``TPROXY_ENABLE_VARDIFF=falseminers in dashboard appear asoffline`.

[warioishere]

Confirmed — the logic looks like what you suspect. Two places both use hashrate !== null as the "miner is online" signal:

• UnifiedDashboard.tsx:99 — activeClients count (shown in the stats header)
• Sv1ClientTable.tsx:68 — the green/grey dot next to each miner name

When enable_vardiff = false, the translator backend doesn't compute hashrate (vardiff is what tracks it), so every client comes back with hashrate: null. The frontend reads that as "disconnected" even though the miners are actively connected.

The correct indicator should be channel_id !== null — a client has a channel when it's actively established, regardless of whether hashrate tracking is on. hashrate should only control what value is displayed in the hashrate column, not whether the dot is green.

I am currently merging the wizard into the UI and could fix this on my presentation next week.

[pavlenex]

Cheers @warioishere let's try to make PR's small in scope (easier to review). So separating these fixed in smaller PR's will make it easier to review. I've also made a proof of concept attempt here #14 as an inspiration.

[GitGab19]

This issue should be properly fixed in the tProxy at the application level.

It's something we mentioned recently during the efforts on stratum-mining/sv2-apps#262 and its PR stratum-mining/sv2-apps#263.

When vardiff is disabled in tProxy, it should be able to update the Sv1 clients hashrate data with what it receives from JDC.

I'll open an issue on sv2-apps and try to tackle it ASAP.

In the meantime, always ensure the enable_vardiff config parameter is always set to TRUE while working on the UI.

[GitGab19]

stratum-mining/sv2-apps#289