Prior to the previous commit, whatever async scheduling occurred caused them to
all have the same tip. Now, some are one block ahead of others. This adds
tolerance for that, as it's an acceptable variance, so long as it's solely one
block.
They used &mut self to prevent execution at the same time. This uses a lock
over the channel to achieve the same security, without requiring a lock over
the entire tributary.
This fixes post-provided Provided transactions. sync_block waited for the TX to
be provided, yet it never would as sync_block held a mutable reference over the
entire Tributary, preventing any other read/write operations of any scope.
A timeout increased (bc2f23f72b) due to this bug
not being identified has been decreased back, thankfully.
Also shims in basic support for Completed, which was the WIP before this bug
was identified.
The Heartbeat was meant to serve for this, yet no Heartbeats are fired when we
don't have active tributaries.
libp2p does offer an explicit KeepAlive protocol, yet it's not recommended in
prod. While this likely has the same pit falls as LibP2p's KeepAlive protocol,
it's at least tailored to our timing.
* dalek 4.0
* cargo update
Moves to a version of Substrate which uses curve25519-dalek 4.0 (not a rc).
Doesn't yet update the repo to curve25519-dalek 4.0 (as a branch does) due
to the official schnorrkel using a conflicting curve25519-dalek. This would
prevent installation of frost-schnorrkel without a patch.
* use half-aggregation for tm messages
* fmt
* fix pr comments
* cargo update
Achieves three notable updates.
1) Resolves RUSTSEC-2022-0093 by updating libp2p-identity.
2) Removes 3 old rand crates via updating ed25519-dalek (a dependency of
libp2p-identity).
3) Sets serde_derive to 1.0.171 via updating to time 0.3.26 which pins at up to
1.0.171.
The last one is the most important. The former two are niceties.
serde_derive, since 1.0.171, ships a non-reproducible binary blob in what's a
complete compromise of supply chain security. This is done in order to reduce
compile times, yet also for the maintainer of serde (dtolnay) to leverage
serde's position as the 8th most downloaded crate to attempt to force changes
to the Rust build pipeline.
While dtolnay's contributions to Rust are respectable, being behind syn, quote,
and proc-macro2 (the top three crates by downloads), along with thiserror,
anyhow, async-trait, and more (I believe also being part of the Rust project),
they have unfortunately decided to refuse to listen to the community on this
issue (or even engage with counter-commentary). Given their political agenda
they seem to try to be accomplishing with force, I'd go as far as to call their
actions terroristic (as they're using the threat of the binary blob as
justification for cargo to ship 'proper' support for binary blobs).
This is arguably representative of dtolnay's past work on watt. watt was a wasm
interpreter to execute a pre-compiled proc macro. This would save the compile
time of proc macros, yet sandbox it so a full binary did not have to be run.
Unfortunately, watt (while decreasing compile times) fails to be a valid
solution to supply chain security (without massive ecosystem changes). It never
implemented reproducible builds for its wasm blobs, and a malicious wasm blob
could still fundamentally compromise a project. The only solution for an end
user to achieve a secure pipeline would be to locally build the project,
verifying the blob aligns, yet doing so would negate all advantages of the
blob.
dtolnay also seems to be giving up their role as a FOSS maintainer given that
serde no longer works in several environments. While FOSS maintainers are not
required to never implement breaking changes, the version number is still 1.0.
While FOSS maintainers are not required to follow semver, releasing a very
notable breaking change *without a new version number* in an ecosystem which
*does follow semver*, then refusing to acknowledge bugs as bugs with their work
does meet my personal definition of "not actively maintaining their existing
work". Maintenance would be to fix bugs, not introduce and ignore.
For now, serde > 1.0.171 has been banned. In the future, we may host a fork
without the blobs (yet with the patches). It may be necessary to ban all of
dtolnay's maintained crates, if they continue to force their agenda as such,
yet I hope this may be resolved within the next week or so.
Sources:
https://github.com/serde-rs/serde/issues/2538 - Binary blob discussion
This includes several reports of various workflows being broken.
https://github.com/serde-rs/serde/issues/2538#issuecomment-1682519944
dtolnay commenting that security should be resolved via Rust toolchain edits,
not via their own work being secure. This is why I say they're trying to
leverage serde in a political game.
https://github.com/serde-rs/serde/issues/2526 - Usage via git broken
dtolnay explicitly asks the submitting user if they'd be willing to advocate
for changes to Rust rather than actually fix the issue they created. This is
further political arm wrestling.
https://github.com/serde-rs/serde/issues/2530 - Usage via Bazel broken
https://github.com/serde-rs/serde/issues/2575 - Unverifiable binary blob
https://github.com/dtolnay/watt - dtolnay's prior work on precompilation
* add Rs() api to SchnorrAggregate
* Correct serai-processor-tests to dalek 4
* fmt + deny
* Slash malevolent validators (#294)
* add slash tx
* ignore unsigned tx replays
* verify that provided evidence is valid
* fix clippy + fmt
* move application tx handling to another module
* partially handle the tendermint txs
* fix pr comments
* support unsigned app txs
* add slash target to the votes
* enforce provided, unsigned, signed tx ordering within a block
* bug fixes
* add unit test for tendermint txs
* bug fixes
* update tests for tendermint txs
* add tx ordering test
* tidy up tx ordering test
* cargo +nightly fmt
* Misc fixes from rebasing
* Finish resolving clippy
* Remove sha3 from tendermint-machine
* Resolve a DoS in SlashEvidence's read
Also moves Evidence from Vec<Message> to (Message, Option<Message>). That
should meet all requirements while being a bit safer.
* Make lazy_static a dev-depend for tributary
* Various small tweaks
One use of sort was inefficient, sorting unsigned || signed when unsigned was
already properly sorted. Given how the unsigned TXs were given a nonce of 0, an
unstable sort may swap places with an unsigned TX and a signed TX with a nonce
of 0 (leading to a faulty block).
The extra protection added here sorts signed, then concats.
* Fix Tributary tests I broke, start review on tendermint/tx.rs
* Finish reviewing everything outside tests and empty_signature
* Remove empty_signature
empty_signature led to corrupted local state histories. Unfortunately, the API
is only sane with a signature.
We now use the actual signature, which risks creating a signature over a
malicious message if we have ever have an invariant producing malicious
messages. Prior, we only signed the message after the local machine confirmed
it was okay per the local view of consensus.
This is tolerated/preferred over a corrupt state history since production of
such messages is already an invariant. TODOs are added to make handling of this
theoretical invariant further robust.
* Remove async_sequential for tokio::test
There was no competition for resources forcing them to be run sequentially.
* Modify block order test to be statistically significant without multiple runs
* Clean tests
---------
Co-authored-by: Luke Parker <lukeparker5132@gmail.com>
* Add DSTs to Tributary TX sig_hash functions
Prevents conflicts with other systems/other parts of the Tributary.
---------
Co-authored-by: Luke Parker <lukeparker5132@gmail.com>
* add slash tx
* ignore unsigned tx replays
* verify that provided evidence is valid
* fix clippy + fmt
* move application tx handling to another module
* partially handle the tendermint txs
* fix pr comments
* support unsigned app txs
* add slash target to the votes
* enforce provided, unsigned, signed tx ordering within a block
* bug fixes
* add unit test for tendermint txs
* bug fixes
* update tests for tendermint txs
* add tx ordering test
* tidy up tx ordering test
* cargo +nightly fmt
* Misc fixes from rebasing
* Finish resolving clippy
* Remove sha3 from tendermint-machine
* Resolve a DoS in SlashEvidence's read
Also moves Evidence from Vec<Message> to (Message, Option<Message>). That
should meet all requirements while being a bit safer.
* Make lazy_static a dev-depend for tributary
* Various small tweaks
One use of sort was inefficient, sorting unsigned || signed when unsigned was
already properly sorted. Given how the unsigned TXs were given a nonce of 0, an
unstable sort may swap places with an unsigned TX and a signed TX with a nonce
of 0 (leading to a faulty block).
The extra protection added here sorts signed, then concats.
* Fix Tributary tests I broke, start review on tendermint/tx.rs
* Finish reviewing everything outside tests and empty_signature
* Remove empty_signature
empty_signature led to corrupted local state histories. Unfortunately, the API
is only sane with a signature.
We now use the actual signature, which risks creating a signature over a
malicious message if we have ever have an invariant producing malicious
messages. Prior, we only signed the message after the local machine confirmed
it was okay per the local view of consensus.
This is tolerated/preferred over a corrupt state history since production of
such messages is already an invariant. TODOs are added to make handling of this
theoretical invariant further robust.
* Remove async_sequential for tokio::test
There was no competition for resources forcing them to be run sequentially.
* Modify block order test to be statistically significant without multiple runs
* Clean tests
---------
Co-authored-by: Luke Parker <lukeparker5132@gmail.com>
* restrict batch size to ~25kb
* add batch size check to node
* rate limit batches to 1 per serai block
* add support for multiple batches for block
* fix review comments
* Misc fixes
Doesn't yet update tests/processor until data flow is inspected.
* Move the block from SignId to ProcessorMessage::BatchPreprocesses
* Misc clean up
---------
Co-authored-by: Luke Parker <lukeparker5132@gmail.com>
For logging purposes, I added code to handle negative time till start. I forgot
to only sleep on positive time till start.
Should fix the recent CI failure.
It was improperly implemented, as it assumed rounds had a constant time
interval, which they do not. It also is against the spec and was meant to
absolve us of issues with poor performance when post-starting blockchains. The
new, and much more proper, workaround for the latter is a 120-second delay
between the Substrate time and the Tributary start time.
By default, tokio-spawned worker panics will only kill the task, not the
program. Due to our extensive use of panicking on invariants, we should ensure
the program exits.
It's largely unoptimized, and not yet exclusive to validators, yet has basic
sanity (using message content for ID instead of sender + index).
Fixes bugs as found. Notably, we used a time in milliseconds where the
Tributary expected seconds.
Also has Tributary::new jump to the presumed round number. This reduces slashes
when starting new chains (whose times will be before the current time) and was
the only way I was able to observe successful confirmations given current
surrounding infrastructure.
The Processor's coins folder referred to the networks it could process, as did
its Coin trait. This, and other similar cases throughout the codebase, have now
been corrected.
Also corrects dated documentation for a key pair is confirmed under the
validator-sets pallet.
This probably should be done with n-long lived tasks, one per Tributary. While
this may not be suitably performant long-term (potential DoS vector), this at
least resolves the halting concerns.
Reduces lock contention.
Additionally changes block_key to include the genesis. While not technically
needed, the lack of genesis introduced a side effect where any Tributary on the
the database could return the block of any other Tributary. While that wasn't a
security issue, returning it suggested it was on-chain when it wasn't. This may
have been usable to create issues.
We had a race condition where'd we be informed of blocks 1 .. 3, and
immediately add 1 .. 3. Because we immediately tried to add 2 after 1, it'd
fail since the tip was still the genesis, yet 2 needs the tip to be 1.
Adding a channel, while ugly, was the simplest way to accomplish this.
Also has any added block be broadcasted. Else there's a race condition where a
node which syncs up to the most recent block does so, yet fails to add the next
block when it's committed to.
This defines the tart of a very complex series of locks I'm really unhappy
with. At the same time, there's not immediately a better solution. This also
should work without issue.
add_active_tributary writes the spec to disk before it returns, so even if the
VecDeque it pushes to isn't popped, the tributary will still be loaded on boot.
Removes last_block as an argument from Tendermint. It now loads from the DB as
needed. While slightly less performant, it's easiest and should be fine.
Impls a LocalP2p for testing.
Moves rebroadcasting into Tendermint, since it's what knows if a message is
fully valid + original.
Removes TributarySpec::validators() HashMap, as its non-determinism caused
different instances to have different round robin schedules. It was already
prior moved to a Vec for this issue, so I'm unsure why this remnant existed.
Also renames the GH no-std workflow from the prior commit.
Necessary as our Tributary chains needed to agree when a Serai block has
occurred, and when a Monero block has occurred. Since those could happen at the
same time, some validators may put SeraiBlock before ExternalBlock and vice
versa, causing a chain halt. Now they can have distinct ordering queues.
The existing code was almost entirely applicable. It just needed to be scoped
with an ID. While the handle function is now a bit convoluted, I don't see a
better option.
[0; 32] is a magic for no block has been set yet due to this being the first
key pair. If [0; 32] is the latest finalized block, the processor determines
an activation block based on timestamps.
This doesn't use an Option for ergonomic reasons.
It originally wasn't an enum so software which had yet to update before an
integration wouldn't error (as now enums are strictly typed). The strict typing
is preferable though.
SubstrateBlock's provision of the most recently acknowledged block has
equivalent information with the same latency. Accordingly, there's no need for
it.
Step moved a step forward after an externally synced/added block. This created
a race condition to add the block between the sync process and the Tendermint
machine. Now that the block routes through Tendermint, there is no such race
condition.
Previously, Tendermint needed to be live more than it needed to be correct.
Under the original intention for it, correctness would fail if any coin
desynced, which would cause the node to fail entirely. By accepting a
supermajority's view of state, despite its own, a single coin's failure would
only lead to inability to participate with that single coin.
Now that Tendermint is solely for Tributary, nodes should halt a coin-specific
chain if their view of the chain differs. They are unable to meaningless
participate regardless.
This also means a supermajority of validators can no longer fake messages from
other validators, allowing the Tributary chain to use uniform weights with much
less impact. There is still enough impact they can't be used (ability to cause
a fork), yet they should allow uniform block production (as that's solely a DoS
concern).
While we prior could've simply additionally checked signatures, add_block's
lack of a failure case would've meant it had to panic. This would've been a DoS
possible a minority-weight *which affected the entire coordinator* and
therefore *the entire validator for all coins*.