hbbft/src/dynamic_honey_badger/dynamic_honey_badger.rs

555 lines
22 KiB
Rust
Raw Permalink Normal View History

use std::collections::BTreeMap;
use std::sync::Arc;
2018-10-25 08:07:52 -07:00
use std::{fmt, result};
use crate::crypto::{PublicKey, SecretKey, Signature};
use bincode;
2018-10-29 07:36:56 -07:00
use derivative::Derivative;
use log::debug;
use rand::Rng;
2018-10-24 02:38:14 -07:00
use serde::{de::DeserializeOwned, Serialize};
use super::votes::{SignedVote, VoteCounter};
use super::{
2018-12-18 07:17:46 -08:00
Batch, Change, ChangeState, DynamicHoneyBadgerBuilder, EncryptionSchedule, Error, FaultKind,
Input, InternalContrib, JoinPlan, KeyGenMessage, KeyGenState, Message, Params, Result,
SignedKeyGenMsg, Step,
};
2018-12-18 07:17:46 -08:00
use crate::fault_log::{Fault, FaultLog};
use crate::honey_badger::{self, HoneyBadger, Message as HbMessage};
use crate::sync_key_gen::{Ack, AckOutcome, Part, PartOutcome, SyncKeyGen};
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
use crate::util;
use crate::{Contribution, DistAlgorithm, Epoched, NetworkInfo, NodeIdT, Target};
/// A Honey Badger instance that can handle adding and removing nodes.
#[derive(Derivative)]
#[derivative(Debug)]
pub struct DynamicHoneyBadger<C, N: Ord> {
/// Shared network data.
pub(super) netinfo: NetworkInfo<N>,
/// The maximum number of future epochs for which we handle messages simultaneously.
2018-10-25 08:07:52 -07:00
pub(super) max_future_epochs: u64,
/// The first epoch after the latest node change.
2018-10-25 08:07:52 -07:00
pub(super) era: u64,
/// The buffer and counter for the pending and committed change votes.
pub(super) vote_counter: VoteCounter<N>,
/// Pending node transactions that we will propose in the next epoch.
pub(super) key_gen_msg_buffer: Vec<SignedKeyGenMsg<N>>,
/// The `HoneyBadger` instance with the current set of nodes.
pub(super) honey_badger: HoneyBadger<InternalContrib<C, N>, N>,
/// The current key generation process, and the change it applies to.
pub(super) key_gen_state: Option<KeyGenState<N>>,
Better proptest persistence through deterministic randomness. (#248) * Add support for RNG instantiation in proptests. * Use `proptest` module strategy to create the rng for `net_dynamic_honey_badger`. * Use seed generation instead of RNG instantiation in tests. * Remove fixed RNG in `generate_map`. * `VirtualNet` now supports setting the random generator through the builder. * Add missing `time_limit` field to `::std::fmt::Debug` trait implementation on `NetBuilder`. * Pass an instantiated random number generator through `NewNodeInfo` as a convenience. * Make the random number generator of `DynamicHoneyBadgerBuilder` configurable, at the cost of now requiring mutability to call `build_first_node()`. * Ensure RNGs are derive from passed in seed in `net_dynamic_hb` tests. * Correct inappropriate use of `random::Random` instead of `Rng::gen` to generate dependent values in `binary_agreement`. The original implementation used `rand::random()`, which will always use the `thread_rng`, ignoring the fact that an RNG has actually been passed in. * Do not use `OsRng` but passed in RNG instead. * Use reference/non-reference passing of rngs more in line with the `rand` crates conventions. * Document `rng` field on `DynamicHoneyBadger`. * Make `SyncKeyGen` work with the extend (`encrypt_with_rng`) API of `threshold_crypto`. * Use passed-in random number generator in `HoneyBadger`. * Create `SubRng` crate in new `util` module to replace `create_rng()`. * Use an RNG seeded from the configure RNG when reinitializing `DynamicHoneyBadger`. * Use the correct branch of `threshold_crypto` with support for passing RNGs.
2018-10-02 07:24:51 -07:00
}
impl<C, N> DistAlgorithm for DynamicHoneyBadger<C, N>
where
2018-10-24 02:38:14 -07:00
C: Contribution + Serialize + DeserializeOwned,
N: NodeIdT + Serialize + DeserializeOwned,
{
2018-08-29 09:08:35 -07:00
type NodeId = N;
type Input = Input<C, N>;
type Output = Batch<C, N>;
type Message = Message<N>;
type Error = Error;
2018-12-18 07:17:46 -08:00
type FaultKind = FaultKind;
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
fn handle_input<R: Rng>(&mut self, input: Self::Input, rng: &mut R) -> Result<Step<C, N>> {
// User contributions are forwarded to `HoneyBadger` right away. Votes are signed and
// broadcast.
match input {
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
Input::User(contrib) => self.propose(contrib, rng),
Input::Change(change) => self.vote_for(change),
}
}
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
fn handle_message<R: Rng>(
&mut self,
sender_id: &Self::NodeId,
msg: Self::Message,
rng: &mut R,
) -> Result<Step<C, N>> {
self.handle_message(sender_id, msg, rng)
}
fn terminated(&self) -> bool {
false
}
fn our_id(&self) -> &N {
2018-08-29 09:08:35 -07:00
self.netinfo.our_id()
}
}
impl<C, N> DynamicHoneyBadger<C, N>
where
2018-10-24 02:38:14 -07:00
C: Contribution + Serialize + DeserializeOwned,
N: NodeIdT + Serialize + DeserializeOwned,
{
/// Returns a new `DynamicHoneyBadgerBuilder`.
pub fn builder() -> DynamicHoneyBadgerBuilder<C, N> {
DynamicHoneyBadgerBuilder::new()
}
/// Creates a new `DynamicHoneyBadger` ready to join the network specified in the `JoinPlan`.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
pub fn new_joining<R: Rng>(
our_id: N,
secret_key: SecretKey,
join_plan: JoinPlan<N>,
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
rng: &mut R,
) -> Result<(Self, Step<C, N>)> {
let netinfo = NetworkInfo::new(
our_id,
None,
join_plan.pub_key_set,
secret_key,
join_plan.pub_keys,
);
let max_future_epochs = join_plan.params.max_future_epochs;
let arc_netinfo = Arc::new(netinfo.clone());
let honey_badger = HoneyBadger::builder(arc_netinfo.clone())
.session_id(join_plan.era)
.params(join_plan.params)
.build();
let mut dhb = DynamicHoneyBadger {
netinfo,
max_future_epochs,
era: join_plan.era,
vote_counter: VoteCounter::new(arc_netinfo, join_plan.era),
key_gen_msg_buffer: Vec::new(),
honey_badger,
key_gen_state: None,
};
let step = match join_plan.change {
ChangeState::InProgress(ref change) => match change {
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
Change::NodeChange(change) => dhb.update_key_gen(join_plan.era, change, rng)?,
_ => Step::default(),
},
ChangeState::None | ChangeState::Complete(..) => Step::default(),
};
Ok((dhb, step))
}
/// Returns `true` if input for the current epoch has already been provided.
pub fn has_input(&self) -> bool {
self.honey_badger.has_input()
}
/// Proposes a contribution in the current epoch.
///
/// Returns an error if we already made a proposal in this epoch.
///
/// If we are the only validator, this will immediately output a batch, containing our
/// proposal.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
pub fn propose<R: Rng>(&mut self, contrib: C, rng: &mut R) -> Result<Step<C, N>> {
2018-10-22 07:09:00 -07:00
let key_gen_messages = self
.key_gen_msg_buffer
.iter()
2018-10-25 08:07:52 -07:00
.filter(|kg_msg| kg_msg.era() == self.era)
2018-10-22 07:09:00 -07:00
.cloned()
.collect();
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
let contrib = InternalContrib {
contrib,
key_gen_messages,
votes: self.vote_counter.pending_votes().cloned().collect(),
};
let step = self
.honey_badger
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
.propose(&contrib, rng)
.map_err(Error::ProposeHoneyBadger)?;
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
self.process_output(step, rng)
}
/// Casts a vote to change the set of validators or parameters.
///
/// This stores a pending vote for the change. It will be included in some future batch, and
/// once enough validators have been voted for the same change, it will take effect.
pub fn vote_for(&mut self, change: Change<N>) -> Result<Step<C, N>> {
if !self.netinfo.is_validator() {
return Ok(Step::default()); // TODO: Return an error?
}
let signed_vote = self.vote_counter.sign_vote_for(change)?.clone();
let msg = Message::SignedVote(signed_vote);
Ok(Target::All.message(msg).into())
}
/// Casts a vote to add a node as a validator.
///
/// This stores a pending vote for the change. It will be included in some future batch, and
/// once enough validators have been voted for the same change, it will take effect.
pub fn vote_to_add(&mut self, node_id: N, pub_key: PublicKey) -> Result<Step<C, N>> {
let mut pub_keys = self.netinfo.public_key_map().clone();
pub_keys.insert(node_id, pub_key);
self.vote_for(Change::NodeChange(pub_keys))
}
/// Casts a vote to demote a validator to observer.
///
/// This stores a pending vote for the change. It will be included in some future batch, and
/// once enough validators have been voted for the same change, it will take effect.
pub fn vote_to_remove(&mut self, node_id: &N) -> Result<Step<C, N>> {
let mut pub_keys = self.netinfo.public_key_map().clone();
pub_keys.remove(node_id);
self.vote_for(Change::NodeChange(pub_keys))
}
/// Handles a message received from `sender_id`.
///
/// This must be called with every message we receive from another node.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
pub fn handle_message<R: Rng>(
&mut self,
sender_id: &N,
message: Message<N>,
rng: &mut R,
) -> Result<Step<C, N>> {
2018-10-25 08:07:52 -07:00
if message.era() == self.era {
match message {
Message::HoneyBadger(_, hb_msg) => {
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
self.handle_honey_badger_message(sender_id, hb_msg, rng)
}
Message::KeyGen(_, kg_msg, sig) => self
.handle_key_gen_message(sender_id, kg_msg, *sig)
.map(FaultLog::into),
Message::SignedVote(signed_vote) => self
.vote_counter
.add_pending_vote(sender_id, signed_vote)
.map(FaultLog::into),
}
2018-10-25 08:07:52 -07:00
} else if message.era() > self.era {
Ok(Fault::new(sender_id.clone(), FaultKind::UnexpectedDhbMessageEra).into())
} else {
// The message is late; discard it.
Ok(Step::default())
}
}
/// Returns the information about the node IDs in the network, and the cryptographic keys.
pub fn netinfo(&self) -> &NetworkInfo<N> {
&self.netinfo
}
/// Returns `true` if we should make our contribution for the next epoch, even if we don't have
/// content ourselves, to avoid stalling the network.
///
/// By proposing only if this returns `true`, you can prevent an adversary from making the
/// network output empty baches indefinitely, but it also means that the network won't advance
/// if fewer than _f + 1_ nodes have pending contributions.
pub fn should_propose(&self) -> bool {
if self.has_input() {
return false; // We have already proposed.
}
if self.honey_badger.received_proposals() > self.netinfo.num_faulty() {
return true; // At least one correct node wants to move on to the next epoch.
}
let is_our_vote = |signed_vote: &SignedVote<_>| signed_vote.voter() == self.our_id();
if self.vote_counter.pending_votes().any(is_our_vote) {
return true; // We have pending input to vote for a validator change.
}
// If we have a pending key gen message, we should propose.
!self.key_gen_msg_buffer.is_empty()
}
/// The epoch of the next batch that will be output.
pub fn next_epoch(&self) -> u64 {
self.era + self.honey_badger.next_epoch()
}
/// Handles a message for the `HoneyBadger` instance.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
fn handle_honey_badger_message<R: Rng>(
&mut self,
sender_id: &N,
message: HbMessage<N>,
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
rng: &mut R,
) -> Result<Step<C, N>> {
if !self.netinfo.is_node_validator(sender_id) {
return Err(Error::UnknownSender);
}
// Handle the message.
let step = self
.honey_badger
.handle_message(sender_id, message)
.map_err(Error::HandleHoneyBadgerMessage)?;
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
self.process_output(step, rng)
}
/// Handles a vote or key generation message and tries to commit it as a transaction. These
/// messages are only handled once they appear in a batch output from Honey Badger.
fn handle_key_gen_message(
&mut self,
sender_id: &N,
kg_msg: KeyGenMessage,
sig: Signature,
2018-12-18 07:17:46 -08:00
) -> Result<FaultLog<N, FaultKind>> {
if !self.verify_signature(sender_id, &sig, &kg_msg)? {
let fault_kind = FaultKind::InvalidKeyGenMessageSignature;
return Ok(Fault::new(sender_id.clone(), fault_kind).into());
}
let kgs = match self.key_gen_state {
Some(ref mut kgs) => kgs,
None => {
return Ok(Fault::new(sender_id.clone(), FaultKind::UnexpectedKeyGenMessage).into());
}
};
// If the sender is correct, it will send at most _N + 1_ key generation messages:
// one `Part`, and for each validator an `Ack`. _N_ is the node number _after_ the change.
if kgs.count_messages(sender_id) > kgs.key_gen.num_nodes() + 1 {
let fault_kind = FaultKind::TooManyKeyGenMessages;
return Ok(Fault::new(sender_id.clone(), fault_kind).into());
}
2018-10-25 08:07:52 -07:00
let tx = SignedKeyGenMsg(self.era, sender_id.clone(), kg_msg, sig);
self.key_gen_msg_buffer.push(tx);
Ok(FaultLog::default())
}
/// Processes all pending batches output by Honey Badger.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
fn process_output<R: Rng>(
&mut self,
hb_step: honey_badger::Step<InternalContrib<C, N>, N>,
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
rng: &mut R,
) -> Result<Step<C, N>> {
let mut step: Step<C, N> = Step::default();
2018-12-18 07:17:46 -08:00
let output = step.extend_with(hb_step, FaultKind::HbFault, |hb_msg| {
Message::HoneyBadger(self.era, hb_msg)
});
for hb_batch in output {
2018-10-25 08:07:52 -07:00
let batch_era = self.era;
2018-11-01 11:18:08 -07:00
let batch_epoch = hb_batch.epoch + batch_era;
let mut batch_contributions = BTreeMap::new();
// Add the user transactions to `batch` and handle votes and DKG messages.
for (id, int_contrib) in hb_batch.contributions {
let InternalContrib {
votes,
key_gen_messages,
contrib,
} = int_contrib;
step.fault_log
.extend(self.vote_counter.add_committed_votes(&id, votes)?);
batch_contributions.insert(id.clone(), contrib);
self.key_gen_msg_buffer
.retain(|skgm| !key_gen_messages.contains(skgm));
2018-10-25 08:07:52 -07:00
for SignedKeyGenMsg(era, s_id, kg_msg, sig) in key_gen_messages {
if era != self.era {
let fault_kind = FaultKind::InvalidKeyGenMessageEra;
2018-10-22 07:09:00 -07:00
step.fault_log.append(id.clone(), fault_kind);
} else if !self.verify_signature(&s_id, &sig, &kg_msg)? {
let fault_kind = FaultKind::InvalidKeyGenMessageSignature;
step.fault_log.append(id.clone(), fault_kind);
2018-10-22 07:09:00 -07:00
} else {
step.extend(match kg_msg {
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
KeyGenMessage::Part(part) => self.handle_part(&s_id, part, rng)?,
2018-10-24 12:23:54 -07:00
KeyGenMessage::Ack(ack) => self.handle_ack(&s_id, ack)?,
2018-10-22 07:09:00 -07:00
});
}
}
}
let change = if let Some(kgs) = self.take_ready_key_gen() {
// If DKG completed, apply the change, restart Honey Badger, and inform the user.
debug!("{}: DKG for complete for: {:?}", self, kgs.public_keys());
self.netinfo = kgs.key_gen.into_network_info().map_err(Error::SyncKeyGen)?;
let params = self.honey_badger.params().clone();
self.restart_honey_badger(batch_epoch + 1, params);
ChangeState::Complete(Change::NodeChange(self.netinfo.public_key_map().clone()))
} else if let Some(change) = self.vote_counter.compute_winner().cloned() {
// If there is a new change, restart DKG. Inform the user about the current change.
match change {
Change::NodeChange(ref pub_keys) => {
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
step.extend(self.update_key_gen(batch_epoch + 1, pub_keys, rng)?);
}
Change::EncryptionSchedule(schedule) => {
self.update_encryption_schedule(batch_epoch + 1, schedule);
2018-10-23 22:18:18 -07:00
}
}
match change {
Change::NodeChange(_) => ChangeState::InProgress(change),
Change::EncryptionSchedule(_) => ChangeState::Complete(change),
}
} else {
ChangeState::None
};
step.output.push(Batch {
2018-11-01 11:18:08 -07:00
epoch: batch_epoch,
2018-10-25 08:07:52 -07:00
era: batch_era,
change,
netinfo: Arc::new(self.netinfo.clone()),
contributions: batch_contributions,
params: self.honey_badger.params().clone(),
});
}
Ok(step)
}
/// Restarts Honey Badger with the new encryption schedule.
pub(super) fn update_encryption_schedule(&mut self, era: u64, schedule: EncryptionSchedule) {
let mut params = self.honey_badger.params().clone();
params.encryption_schedule = schedule;
self.restart_honey_badger(era, params);
}
/// If the winner of the vote has changed, restarts Key Generation for the set of nodes implied
/// by the current change.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
pub(super) fn update_key_gen<R: Rng>(
2018-10-23 22:18:18 -07:00
&mut self,
2018-10-25 08:07:52 -07:00
era: u64,
pub_keys: &BTreeMap<N, PublicKey>,
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
rng: &mut R,
2018-10-23 22:18:18 -07:00
) -> Result<Step<C, N>> {
if self.key_gen_state.as_ref().map(KeyGenState::public_keys) == Some(pub_keys) {
return Ok(Step::default()); // The change is the same as before. Continue DKG as is.
}
debug!("{}: Restarting DKG for {:?}.", self, pub_keys);
let params = self.honey_badger.params().clone();
self.restart_honey_badger(era, params);
let threshold = util::max_faulty(pub_keys.len());
let sk = self.netinfo.secret_key().clone();
2018-08-29 09:08:35 -07:00
let our_id = self.our_id().clone();
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
let (key_gen, part) = SyncKeyGen::new(our_id, sk, pub_keys.clone(), threshold, rng)
.map_err(Error::SyncKeyGen)?;
self.key_gen_state = Some(KeyGenState::new(key_gen));
if let Some(part) = part {
self.send_transaction(KeyGenMessage::Part(part))
} else {
Ok(Step::default())
}
}
/// Starts a new `HoneyBadger` instance and resets the vote counter.
fn restart_honey_badger(&mut self, era: u64, params: Params) {
2018-10-25 08:07:52 -07:00
self.era = era;
self.key_gen_msg_buffer.retain(|kg_msg| kg_msg.0 >= era);
let netinfo = Arc::new(self.netinfo.clone());
2018-10-25 08:07:52 -07:00
self.vote_counter = VoteCounter::new(netinfo.clone(), era);
self.honey_badger = HoneyBadger::builder(netinfo)
2018-10-25 08:07:52 -07:00
.session_id(era)
.params(params)
.build();
}
/// Handles a `Part` message that was output by Honey Badger.
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
fn handle_part<R: Rng>(
&mut self,
sender_id: &N,
part: Part,
rng: &mut R,
) -> Result<Step<C, N>> {
2018-10-03 04:21:59 -07:00
let outcome = if let Some(kgs) = self.key_gen_state.as_mut() {
2018-10-24 12:23:54 -07:00
kgs.key_gen
OsRng / external RNG Refactoring (#357) * Use `OsRng` in place of `thread_rng`. This changes the defaults of any builder by instantiating an `OsRng` instead of a `thread_rng`, the former being much more secure than the latter. Additionally, all the unit tests that still instantiate RNGs manually used `OsRng`s as well; while there is no actual need for this level of security in tests, the performance overhead is very small and random number generation complexity has such a small impact on these tests that the convenience of being able to ban `thread_rng` from the codebase altogether, setting a good example and avoid issues when refactoring later greatly outweigh the negatives. * Instead of storing random number generators in the various consensus algorithm instances, pass them in from the outside whenever they are needed. This changes a large amount of interfaces (and in this commit is only partially done, since `DistAlgorithm` needs to be fundamentally altered as well. It also obsoletes parts of the `util` module. * Added an `R: Rng` type parameter to both methods of `DistAlgorithm`, forcing callers to pass in their own Rngs. * Fixed documentation grammar and spelling in some of the altered interfaces due to RNG refactoring. * Move `rng` argument to the end of the argument for most functions. Also includes a reformatting due to Rust 1.30. * Updated tests, accomodate `rng`-API changes. * Fixed remaining compilation issues with new RNG code. * Fix illegal `self` import outside curly braces. * Cleaned up comments and fixed broken definition of `broadcast_input`. * Updated existing test cases to properly work with static dispatch randomness. * Do not use boxed `Rng`s for key generation in test networks. * Use the passed-in `Rng` in `ReorderingAdversary`, instead of storing a boxed one. * Fixed clippy lints after refactoring. * Removed some no-longer necessary manual `fmt::Debug` implementations in test framework. * Use `OsRng` even in tests in `binary_agreement_mitm`. * Use a proper deterministic RNG in tests `binary_agreement_mitm`. * Refactor `examples/simulation.rs` by not using `ThreadRng`, passing generic `Rng` parameters throughout and using a type alias instead of a newtype as the `Transaction`. * Remove `thread_rng` use from `examples/node.rs`. * Explicitly construct `InternalContrib` in `DynamicHoneyBadger::propose`. * Fixed typo in description of `DistAlgorithm` trait.
2018-12-14 04:51:09 -08:00
.handle_part(&sender_id, part, rng)
.map_err(Error::SyncKeyGen)?
2018-10-03 04:21:59 -07:00
} else {
2018-10-22 07:09:00 -07:00
// No key generation ongoing.
let fault_kind = FaultKind::UnexpectedKeyGenPart;
return Ok(Fault::new(sender_id.clone(), fault_kind).into());
Better proptest persistence through deterministic randomness. (#248) * Add support for RNG instantiation in proptests. * Use `proptest` module strategy to create the rng for `net_dynamic_honey_badger`. * Use seed generation instead of RNG instantiation in tests. * Remove fixed RNG in `generate_map`. * `VirtualNet` now supports setting the random generator through the builder. * Add missing `time_limit` field to `::std::fmt::Debug` trait implementation on `NetBuilder`. * Pass an instantiated random number generator through `NewNodeInfo` as a convenience. * Make the random number generator of `DynamicHoneyBadgerBuilder` configurable, at the cost of now requiring mutability to call `build_first_node()`. * Ensure RNGs are derive from passed in seed in `net_dynamic_hb` tests. * Correct inappropriate use of `random::Random` instead of `Rng::gen` to generate dependent values in `binary_agreement`. The original implementation used `rand::random()`, which will always use the `thread_rng`, ignoring the fact that an RNG has actually been passed in. * Do not use `OsRng` but passed in RNG instead. * Use reference/non-reference passing of rngs more in line with the `rand` crates conventions. * Document `rng` field on `DynamicHoneyBadger`. * Make `SyncKeyGen` work with the extend (`encrypt_with_rng`) API of `threshold_crypto`. * Use passed-in random number generator in `HoneyBadger`. * Create `SubRng` crate in new `util` module to replace `create_rng()`. * Use an RNG seeded from the configure RNG when reinitializing `DynamicHoneyBadger`. * Use the correct branch of `threshold_crypto` with support for passing RNGs.
2018-10-02 07:24:51 -07:00
};
2018-10-03 04:21:59 -07:00
match outcome {
2018-10-24 12:23:54 -07:00
PartOutcome::Valid(Some(ack)) => self.send_transaction(KeyGenMessage::Ack(ack)),
PartOutcome::Valid(None) => Ok(Step::default()),
PartOutcome::Invalid(fault) => {
let fault_kind = FaultKind::SyncKeyGenPart(fault);
Ok(Fault::new(sender_id.clone(), fault_kind).into())
}
}
}
/// Handles an `Ack` message that was output by Honey Badger.
2018-10-24 12:23:54 -07:00
fn handle_ack(&mut self, sender_id: &N, ack: Ack) -> Result<Step<C, N>> {
let outcome = if let Some(kgs) = self.key_gen_state.as_mut() {
kgs.key_gen
.handle_ack(sender_id, ack)
.map_err(Error::SyncKeyGen)?
} else {
2018-10-22 07:09:00 -07:00
// No key generation ongoing.
let fault_kind = FaultKind::UnexpectedKeyGenAck;
return Ok(Fault::new(sender_id.clone(), fault_kind).into());
2018-10-24 12:23:54 -07:00
};
match outcome {
AckOutcome::Valid => Ok(Step::default()),
AckOutcome::Invalid(fault) => {
let fault_kind = FaultKind::SyncKeyGenAck(fault);
Ok(Fault::new(sender_id.clone(), fault_kind).into())
}
}
}
/// Signs and sends a `KeyGenMessage` and also tries to commit it.
fn send_transaction(&mut self, kg_msg: KeyGenMessage) -> Result<Step<C, N>> {
let ser = bincode::serialize(&kg_msg).map_err(|err| Error::SerializeKeyGen(*err))?;
let sig = Box::new(self.netinfo.secret_key().sign(ser));
if self.netinfo.is_validator() {
let our_id = self.our_id().clone();
2018-10-25 08:07:52 -07:00
let signed_msg = SignedKeyGenMsg(self.era, our_id, kg_msg.clone(), *sig.clone());
self.key_gen_msg_buffer.push(signed_msg);
}
2018-10-25 08:07:52 -07:00
let msg = Message::KeyGen(self.era, kg_msg, sig);
Ok(Target::All.message(msg).into())
}
/// If the current Key Generation process is ready, returns the `KeyGenState`.
///
/// We require the minimum number of completed proposals (`SyncKeyGen::is_ready`) and if a new
/// node is joining, we require in addition that the new node's proposal is complete. That way
/// the new node knows that it's key is secret, without having to trust any number of nodes.
fn take_ready_key_gen(&mut self) -> Option<KeyGenState<N>> {
if self
.key_gen_state
.as_ref()
.map_or(false, KeyGenState::is_ready)
{
self.key_gen_state.take()
} else {
None
}
}
/// Returns `true` if the signature of `kg_msg` by the node with the specified ID is valid.
/// Returns an error if the payload fails to serialize.
///
/// This accepts signatures from both validators and currently joining candidates, if any.
fn verify_signature(
&self,
node_id: &N,
sig: &Signature,
kg_msg: &KeyGenMessage,
) -> Result<bool> {
let ser = bincode::serialize(kg_msg).map_err(|err| Error::SerializeKeyGen(*err))?;
let verify = |opt_pk: Option<&PublicKey>| opt_pk.map_or(false, |pk| pk.verify(&sig, &ser));
let kgs = self.key_gen_state.as_ref();
let current_key = self.netinfo.public_key(node_id);
let candidate_key = kgs.and_then(|kgs| kgs.public_keys().get(node_id));
Ok(verify(current_key) || verify(candidate_key))
}
2018-10-25 08:07:52 -07:00
/// Returns the maximum future epochs of the Honey Badger algorithm instance.
pub fn max_future_epochs(&self) -> u64 {
self.max_future_epochs
}
}
impl<C, N> fmt::Display for DynamicHoneyBadger<C, N>
where
C: Contribution + Serialize + DeserializeOwned,
N: NodeIdT + Serialize + DeserializeOwned,
{
fn fmt(&self, f: &mut fmt::Formatter) -> result::Result<(), fmt::Error> {
2018-10-25 08:07:52 -07:00
write!(f, "{:?} DHB(era: {})", self.our_id(), self.era)
}
}
impl<C, N> Epoched for DynamicHoneyBadger<C, N>
where
C: Contribution + Serialize + DeserializeOwned,
N: NodeIdT + Serialize + DeserializeOwned,
{
type Epoch = (u64, u64);
fn epoch(&self) -> (u64, u64) {
(self.era, self.honey_badger.epoch())
}
}