search

ethereumproject / ECIPs

The Ethereum Classic Improvement Proposal
55 stars 47 forks source link

Addressing security issues with a January hardfork? #11

Closed arvicco closed 7 years ago

arvicco commented 7 years ago

The latest round of attacks against ETH unearthed a number of possible attack vectors against the network. Some of them may be fixed/patched and thus mitigated, but a hard fork is required for a permanent solution, as discussed here: https://github.com/ethereum/EIPs/issues/150

My question is, should we strive to include such fix in a January hardfork, while we are at it? Would be interested in your perspective.

@whatisgravity @splix @igetgames @elaineo @mikeyb @ericsomdahl @avtarsehra @realcodywburns

elaineo commented 7 years ago

I'm not a huge fan of the strategy that's being used there -- guessing at numbers for gas costs and seeing what happens. If the guesses are bad, we have to HF again. Surely we can come up with a better strategy? (Probably not in time for January though.)

Is it sufficient to set default geth gas limit lower?

trustfarm-dev commented 7 years ago

@elaineo I think make one time trasfer gas fee to low. But, In case of multiple contract loop , eg

  1. for (1 ; xxx ; many) or while (many--)
  2. recursefunc() { recursefunc() }
    • like 1,2 case of codes, gas fee must increase multiple ratio and exponentially, when branch call or goto call, prevent DDOS contract attacks.

It needs more complicated approach, I think.

Recent days of ETH, there's several DDOS Attack vectors, so, most of geth and parity has big difficulties making blocks. theres' many Orphans blocks.