Nicolas Gailly
Nicolas Gailly
For example drand panics when your private listen and control port are the same
Show the command to run a certbot cron to automatically renew TLS certificates from let's encrypt.
We should have a common definition of a beacon and use it everywhere. Currently there is `protocol.proto` that defines `BeaconPacket` and in `api.proto` we have `PublicRandResponse` which are the same.
There is for example full suite of struct implementing `chain,Store` which are located in `chain/beacon/store.go`. I'm pretty sure we can find some other examples of better "packaging" in these.
```go // core/drand_beacon_control.go nodes := group.Nodes if err := bp.pushDKGInfo([]*key.Node{}, nodes, 0, group, in.GetInfo().GetSecret(), in.GetInfo().GetTimeout(), beaconID); err != nil { return nil, err } ``` ```go // core/drand_beacon_control.go func (bp...
Currently in the fast sync mode, if the timer kicks in, the code will re-send the responses or justifications etc regardless of the current state. This has no direct implication...
Specifically the gRPC protobufs services and the DKG structs (lacking signature field)
We need to bring timed encryption to production to be used by end users. That includes: - Polishing up the PoC here https://github.com/drand/kyber/tree/feat/basepoint/encrypt/timelock - Integrate in as a client feature...
Currently the daemon log only prints before signalling the key but does not log any error.
Right now, when there's an error when loading the group/distkey it automatically switch and tries to reset the database and/or else. This path of failure should be more explicit and...