CodeChain-io / CodeChain-io/codechain

OnChain invalid timestamp block propagation fail

Open
#808 1 comment 0 reactions 0 assignees View on GitHub
bug test
Dominant language
Rust
Stars
256
Forks
50
PR merge metrics
No merged PRs in 30d

Description

There are some issues in sending invalid timestamp onChain blocks.
---
Below is a test scenario.
1. Make block1 and block2.
2. The timestamp in Block2's header should be invalid. Simply make it smaller than block1.
2. Send a `status message` with block2's hash as the bestHash.
3. Send a `header response message` which is [genesisHeader, block1Header, block2Header].
4. Send a `body response message` == [[], []], which is correspond to block1's body and block2's body.
5. Block is accepted from the target node only for block1. So the target node should send a status message with block1's hash as bestHash.(This is normal behaviour but it didn't)
---
And here are problems
1. The target node does not send `status message` if in the above situation. However, if I inserted `info!("55");` then the `status message` come properly.
2. In below `verify()` function in `/core/verification/queue/mod.rs`, `Ok(verified)` is returned twice although the block2 is invalid.
```rust

fn verify(
verification: Arc>,
engine: Arc,
ready_signal: Arc,
empty: Arc,
more_to_verify: Arc,
_id: usize,
) {
loop {
// wait for work if empty.
{
let mut more_to_verify_mutex = verification.more_to_verify_mutex.lock().unwrap();

if verification.unverified.lock().is_empty() && verification.verifying.lock().is_empty() {
empty.notify_all();
}

while verification.unverified.lock().is_empty() {
more_to_verify_mutex = more_to_verify.wait(more_to_verify_mutex).unwrap();
}
}

// do work.
let item = {
// acquire these locks before getting the item to verify.
let mut unverified = verification.unverified.lock();
let mut verifying = verification.verifying.lock();

let item = match unverified.pop_front() {
Some(item) => item,
None => continue,
};

verification.sizes.unverified.fetch_sub(item.heap_size_of_children(), AtomicOrdering::SeqCst);
verifying.push_back(Verifying {
hash: item.hash(),
output: None,
});
item
};

let hash = item.hash();
let is_ready = match K::verify(item, &*engine, verification.check_seal) {
Ok(verified) => {
let mut verifying = verification.verifying.lock();
let mut idx = None;
for (i, e) in verifying.iter_mut().enumerate() {
if e.hash == hash {
idx = Some(i);

verification
.sizes
.verifying
.fetch_add(verified.heap_size_of_children(), AtomicOrdering::SeqCst);
e.output = Some(verified);
break
}
}

if idx == Some(0) {
// we're next!
let mut verified = verification.verified.lock();
let mut bad = verification.bad.lock();
VerificationQueue::drain_verifying(
&mut verifying,
&mut verified,
&mut bad,
&verification.sizes,
);
info!("111");
true
} else {
info!("222");
false
}
}
Err(_) => {
let mut verifying = verification.verifying.lock();
let mut verified = verification.verified.lock();
let mut bad = verification.bad.lock();

bad.insert(hash.clone());
verifying.retain(|e| e.hash != hash);

if verifying.front().map_or(false, |x| x.output.is_some()) {
VerificationQueue::drain_verifying(
&mut verifying,
&mut verified,
&mut bad,
&verification.sizes,
);
info!("33");
true
} else {
info!("44");
false
}
}
};
if is_ready {
info!("55");
// Import the block immediately
ready_signal.set_sync();
}
}
}
```

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.