Recovery from unexpectedly large Eth reorg #480
Unanswered
MaksymZavershynskyi
asked this question in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
This issue relates to all components of the bridge, so I have created it in this repo. It is mostly related to Ropsten, but with a very low chance it might be useful for Ethereum Mainnet.
Problem 1
Sometimes Ropsten reorgs thousands of blocks see #329 . If depth of re-org exceeds the depth of
finalized_gc_threshold
inEthOnNearClient
then the bridge will break. Specifically, the client won't be able to switch to the new canonical chain.We need to discuss if there is a DevX-acceptable way to address this problem.
Problem 2
If Ropsten re-orgs only exceed
num_confirmations
but do not exceedfinalized_gc_threshold
(we are transitioning this configuration parameter into connectors near/rainbow-bridge-rs#3) there are two issues:The first issue is not addressable by engineering. The second issue however can be addressed if connectors are somehow equipped to keep history by associated minted or locked assets with the blocks at which they were locked/minted. If these blocks are reverted they should revert the assets too restoring equilibrium. For example:
The connector contract should then recover in the following way:
Beta Was this translation helpful? Give feedback.
All reactions