Do you guys think that the Unavailibility Penalty should be more harsh to the validators who are ruled out of election because they were below 66.66% of signing rate?
Currently, validators can come back as active after 1 epoch (after the one they were ruled out). I personally think that all validators should have a part of responsibility in frequently checking their node setup and if a validator is ruled out, the validator should be penalized by not being able to get back to the pool for X number of epoch. It is way too easy for someone to run a node, leave it and whenever they are ruled out, just to come back, wait 1 day and start being elected again (my opinion).
Also, does missing sign in any way affects the network performance?
In every epoch validators should sign more than 2/3 of the signatures that its BLS keys are asked to sign.
At the end of each epoch, the validators with uptime of no more than 2/3 (66.6%) will have their status set to “Inactive” and be ruled out from the new election. For these inactive validators, they are required to manually set their status to “Active” by sending an EditValidator transaction…
Looking back at some validators who are having issues with signing in current epoch 793
Yeah. I believe could be an escalating thing. I mean a single failure can happen to anyone, anywhere.
But if it becomes a routine, as we experienced at some point with some keyplayers, the ones responsible for it should suffer more heavy consequences.
If it’s feasible for the network to keep track of the times that a validator went unelected due to poor signrate in a time-frame. So, the subsequent unelections due to this same motive, should be heavier. Maybe going geometric. 1 epoch at the first, 2 at the second. 4 at the third and so on.
Also, if a validator that receives this flag/track, could have this status lifted after x epochs without being unelected for the same motive. So you can give a chance to them to redeem themselves.
Penalties make sense for malicious behavior, but in the case when the Validator is not available he will never get elected actually and mostly because no token holder will stake with him.
IMHO a penalty for a low uptime level will create a barrier for new validators - unknown field, no malicious intentions but already threats for his investment.
In kind of there is a penalty, first by bad signing rate you get only rewards for signed blocks and second a Epoch you will earn nothing.
Lately there where even some issues that signing rate dropped even your node was okay. On epoch 789 I got Bingo’s and my signing rate I requested over HMY was 99.5%, Smartstake was offline (may due to the AWS problem?). After waking up I saw in that timeframe I had 0% signing rate and finished the epoch with 66.8%. there was also others with similar issues in the last week’s.
So in that case my node was good and there was some network issue. how will you separated network related drops from human error?
It makes even more sense to have it escalating for recurring on it. Being new and having some erros is ok.
But if you dont address the issue and keep running in the same error, how you would improve? If you are in a comfortable position and have a high selfstake that enable you to be elected, being flushed only sometimes that you can’t contribute to the network will also lift a barrier where another more serious validator could be noticed instead.
We need to see how many selfstakers with such a high amount of staking we have because those who are able to selfstake hundreds of thousands of dollars ( only at this level they can be elected) are doing it for profits for sure and we come back to motivation - they will do all necessary to get their profits ( means keep the node in a proper condition or delegate to someone ).
Any point of view at this stage is just a theoretical discussion - I respect yours , but penalties always are serving as a barrier for inexperienced. And we need them actually if we look to have a serious growth of small validators number.
I´m quite new to the protocol as a validator myself, started at the mid july. But in those almost 4 months, I´ve saw this action happening more than once. By the same guys “eager” to seek profit. I´m not going to point fingers, but you can try to check it via smartstake. Maybe this talk is fruitless, but barriers must be set to avoid malicious intent , as well as to educate.
I agree that we need new validators and many could be inexperienced, I was one of them, spent days learning and trying to setup my node.
But when someone delegates with you, it trusts you with vote power and even more, they trust in your node to bring them profit. You´re providing a service, it´s a serious matter. And by being so, must be taken seriously. Without severe consequencies it could be handled in a frivolous manner.
Mistakes can happen, shortages, VPS cloud could fail, once in a while is ok, and those are not the case to be flagged here. But the recurrence of those “happenings” must be treated with a more severe look. IDK with time frame would be ideal, weekly? Once every 7 epochs? Must be decided by the community, but I strongly agree that there a boundary must be set to separate casual and fortuitous cases from careless or malicious nodes.
