Content
@
0 reply
1 recast
2 reactions
Cas Cusic
@cascusic
@toly Some thoughts after yesterday’s Solana downtime: 1. We were back up pretty dang fast - the social coordination layer allowed this 2. There’s a flip side to this. The majority of validators immediately updated their systems, before the code was thoroughly looked over by outside sources.
2 replies
0 recast
1 reaction
Anatoly Yakovenko
@toly
More clients and more distributions help. Jito and fd don’t blindly rebase
1 reply
0 recast
0 reaction
Cas Cusic
@cascusic
This makes sense. I trust with time then that we will be sufficiently distributed to make this more of a non issue - doesn’t seem like we’re there yet but I definitely see where we are going.
1 reply
0 recast
1 reaction
Anatoly Yakovenko
@toly
Everything being open source would make the attack short lived. So there is some peace there. Although a short lived attack can be quite damaging
2 replies
0 recast
0 reaction
Cas Cusic
@cascusic
The way I see it, I believe the social layer is going to ultimately be turned into the absolute last ditch - especially as we see solana take off into the future. It’s cool now, but as it gets bigger I believe rollbacks will become more and more difficult to coordinate.
1 reply
0 recast
0 reaction
Anatoly Yakovenko
@toly
If social slashing and rollback is impossible to coordinate then the majority can do what it wants. The promise of the revolution can only survive as a perpetual revolution.
1 reply
0 recast
0 reaction
Cas Cusic
@cascusic
I would agree. but am I wrong to believe that with more decentralization in software relating to running nodes, the chances of a chain crash go down? Let me ask you a more direct question - what could we have done differently here? What is going to keep issues like this from shutting down the chain in the future?
1 reply
0 recast
1 reaction
