fix(p2p): validate chain exchange responses for block headers #4454
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
hanabi1224
requested review from
LesnyRumcajs and
elmattic
and removed request for
a team
src/chain_sync/network_context.rs
Outdated
Comment on lines
150
to
167
| |tipsets: &Vec<Arc<Tipset>>| { | ||
| if let Some(start) = tipsets.first() { | ||
| if start.key() != tsk { | ||
| tracing::warn!(epoch=%start.epoch(), expected=%tsk, actual=%start.key(), "start tipset key mismatch"); | ||
| return false; | ||
| } | ||
| for (ts, pts) in tipsets.iter().zip(tipsets.iter().skip(1)) { | ||
| if ts.parents() != pts.key() { | ||
| tracing::warn!(epoch=%ts.epoch(), expected_parent=%pts.key(), actual_parent=%ts.parents(), "invalid chain"); | ||
| return false; | ||
| } | ||
| } | ||
| true | ||
| } else { | ||
| tracing::warn!(%count, "invalid empty chain_exchange_headers response"); | ||
| false | ||
| } | ||
| }, |
There was a problem hiding this comment.
Would it make sense to extract this to a named method a write a unit test to check and document the behaviour? It also might prevent future regressions.
…hain_exchange_headers
LesnyRumcajs
approved these changes
Jun 26, 2024
@LesnyRumcajs unit tests added! |
| @@ -562,4 +597,37 @@ mod tests { | |||
| assert_eq!(batch.get_ok().await, None); | |||
| assert!(exceeded.load(Ordering::Relaxed)); | |||
| } | |||
|
|
|||
| #[test] | |||
| #[allow(unused_variables)] | |||
There was a problem hiding this comment.
What variables are unused?
There was a problem hiding this comment.
genesis_header, first_header etc.
hanabi1224
force-pushed
the
hm/validate-result--chain_exchange_headers
branch
from
c52a1c5
to
0acd8e2
Compare
hanabi1224
commented
Jun 27, 2024
src/libp2p/peer_manager.rs
Outdated
| } | ||
| let cost = if info.successes + info.failures > 0 { |
There was a problem hiding this comment.
wibble: this was changed to if info.successes > 0 to address #4376 (comment)
However, this unexpectedly changes the logic to favor peers that never succeed.
The original code does not panic but returns inf
fn main() {
let x = 100.0 + f64::from(1) / f64::from(0);
println!("{x}");
}
// Output: inf
LesnyRumcajs
approved these changes
Jun 27, 2024
…hain_exchange_headers
…hain_exchange_headers
lemmih
approved these changes
Jul 11, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary of changes
Part of #4441 investigation
Confirmed that
ChainExchangefailures should be in the console output or log. Theoretically, there is a chance that a remote peer always returns a successful empty response and leads the downloading code to a dead loop with no errors.This PR tries to mitigate the issue by adding validation logic to the chain exchange response
Changes introduced in this pull request:
NonEmptyU64instead ofu64forchain_exchangerequest_lenReference issue to close (if applicable)
Closes
Other information and links
Change checklist