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
At the moment ATXs are published immediately after building them leaving the task of validation to go-libp2p validation routines. This has a few disadvantages:
if ATX fails validation libp2p doesn't log why validation failed
the ATX is cached by libp2p when publishing. We had issues in the past where the ATX failed publication once, the node recreated PoST for that ATX and published it again. libp2p then reported back that the publication was successful (since the ATX was already cached) masking the problem and complicating finding the cause of the issue
any ATX not published by an identity managed by the node must be stored when it passed syntactical validation, even if it fails contextual validation, where the node instead should publish a malfeasance proof for the malicious ATX. In contrast to that ATXs published by an identity managed by the node should only be stored when they pass both syntactical and contextual validation and malfeasance proofs should never be published against one self.
Affected code
Code in the activation package. Specifically the Builder and Handler code.
The text was updated successfully, but these errors were encountered:
Description
At the moment ATXs are published immediately after building them leaving the task of validation to go-libp2p validation routines. This has a few disadvantages:
Affected code
Code in the
activation
package. Specifically theBuilder
andHandler
code.The text was updated successfully, but these errors were encountered: