Nonce too low - nonce jumps #624
Assignees
Comments
Merged
mkysel
added a commit
that referenced
this issue
Mar 13, 2025
When hammering our alchemy node in testnet, I noticed a bunch of issues that weren't seen in the local test development. 1) "nonce too high" - it usually indicates that the mempool is too full. - gracefully hide the error from the caller if it does get hit - prevent it from happening as often by limiting the concurrent number of transactions in flight 2) #624 - Added a few tests that are useful, but dont fix 624 3) drive-by fix of TestAddRates Otherwise HA payer works perfectly. Closes #423 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit ## Summary by CodeRabbit - **New Features** - Introduced a delay mechanism that gracefully handles transient operational issues, contributing to smoother transaction processing. - Enhanced concurrent request management to improve performance and stability during simultaneous operations. - **Tests** - Added new tests to verify the resilience and robustness of concurrent operations under load. - Improved error handling in tests to manage transient failures more effectively, including a retry mechanism for certain operations. <!-- end of auto-generated comment: release notes by coderabbit.ai -->
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We might see logs like these
Our nonce manager can handle these gracefully, but it might take some time to get to the current state.
I've seen the nonce to jump 200-300 elements forward.
It would be cool if we could fast forward the nonce manager using
abandon, but it has been super hard to write and I can't get a quick fix.The text was updated successfully, but these errors were encountered: