You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Aug 19, 2022. It is now read-only.
A "design" covering the developer experience should be posted first before we jump into impelementation.
We can only allowlist on IP address (not peer-id) because the accept/reject decision happens before we read enough of the stream to determine peer id.
An eclipse attack would be mitigated by setting a default incoming connection limit to 0, but then ALLOW-listing certain trusted IP addresses.
Basic static allow-listing can also be accomplished at the OS level with "firewall rules" that DENY all incoming requests but ALLOW requests for certain trusted IPs. The advantage of doing this in libp2p itself is it makes it easy to build an intelligent application. For example, an application could observe that certain nodes have been reputable for a sustained period of time and thus move them to the trusted set of "ALLOW list" IP addresses so already have this set even if an eclipse attack occurs and default incoming limits need to be set to zero.
Done Criteria
Why Important
This can be a useful mechanism for protecting against eclipse attacks.
Notes
The text was updated successfully, but these errors were encountered: