NethermindEth / NethermindEth/Catalyst
Better recovery from `preconfBlock` timeout
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 44
- Forks
- 18
- PR merge metrics
- No merged PRs in 30d
Description
Right now, when we get Failed to call driver RPC for API 'preconfBlocks' within the duration (1000ms), we remove the block that failed to preconfBlock like this:
match self
.taiko
.advance_head_to_new_l2_block(
l2_block,
anchor_block_id,
l2_slot_info,
end_of_sequencing,
)
.await
{
Ok(preconfed_block) => Ok(preconfed_block),
Err(err) => {
error!("Failed to advance head to new L2 block: {}", err);
self.remove_last_l2_block().await?;
Ok(None)
}
}
However, this sometimes causes the node to be stuck as the preconfBlock may actually be reflected in taiko geth, but removed from side car. This causes the side car to be stuck with highestUnsafeL2PayloadBlockID: 407, different from Taiko Geth Height: 408.
We should handle recovery from such time outs in a better way.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the shown preconfBlock timeout handling and the call to remove_last_l2_block(). Trace how the sidecar state is reconciled with Taiko Geth after the timeout, especially the highestUnsafeL2PayloadBlockID mismatch. Done means a timeout recovery path no longer leaves the sidecar behind when the preconfBlock was accepted by Taiko Geth.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- blockchain, distributed-systems
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100