Made the disqualification temporarily

There are situations when a node with one sattelite is disqualified. It continues to work with other sattelites. I have this problem. Why not make the disqualification temporary, so that in a month or 2 you can recover with an impeccable reputation. Send only audits, if everything is good - restore. It doesn't make sense to create a new node and start all over again if the disqualifications are permanent, because they will continue. But for honest and efficient nodes to make a chance to resume work. If the light is suddenly turned off and data is lost, then pay the operator for it? I personally did not know that you need to disable the write cache on the disk and paid for this eternal disqualification. Why such strict requirements? It is not always possible to perform them flawlessly. Yes, even the list of necessary actions before starting the node is not written anywhere, but for this they are disqualified. Maybe you should give a small chance to resume working with sattelite, if it happened only once?

Writed with translator. 

  • Dmitrii Sergeev
  • Jan 15 2020
  • Attach files
  • Pierre-Antoine Charrier commented
    15 Jan 22:39

    More globally I think that nodes should be fully automated. Disqualification shouldn't be a thing unless it's evident that the node operator is trying to cheat. But I don't want my node to be disqualified because of a power outage during my 2 week hollydays for instance.

    It would make sense to reactivate some escrow mechanisms if this were to happen, or to earn way less money while the network is rebuilding the data that became unavailable. But really, disqualification should happen in last resort, if the node is clearly bad, cheating, or has been away for months maybe (but even in this case, if it were to get back online eventually, it could simply start back with 0 data and reputation).

  • Florian Lhermite commented
    09 Mar 05:20

    Google permits to fill a form and you have a last chance to recover your disqualified account. I think a simple form (maybe included in the node/dashboard when it happens, or a generated link to the form) would be great.

    And/Or maybe 3 alerts before definitive disqualification (like the 3 strikes on Youtube), for the lifetime of the node (or one full year before it resets again, to decide). So the counter automatically increases of 1 when the uptime/audit is below the minimum required. Then, you have to recover the minimum before a fixed dead end datetime (which can be computed easily to define a good estimation to recover). If it persists, it increments to 2, and you've again a last chance to recover it. Last one is definitive. In that case, if the SNO is well notified by the warning, he has the ability to react at time to avoid the disqualification. When alert is reached, satellittes can take measures to repair data, if the failing audit/uptime are critical, and deleting repaired data/copied data if the node recovers...