Open
Description
Description
I had a couple of ideas for additional anti-slashing measures. I’ve seen tooling that monitors a set of validators and kills VC processes if one of those validators get slashed. I believe we could implement similar functionality.
Our ValidatorMonitor
already monitors slashings for a user-selected set of validators. We could extend this is one or more of the following ways:
- Stop producing attestations and blocks if one of those validators gets slashed.
- Open up a new endpoint on the BN that indicates if one of those has been slashed. The VC polls that endpoint and shuts down if there has been a slashing. This could be a candidate for a standard API, all BNs could watch for slashings on VCs that have used validator registration APIs.
In both of these cases I’d suggest writing a file to disk to prevent future startups. The user would manually delete that file once they’re confident the event is resolved.