EOS EVM Node v0.7.0-rc2
Pre-release
This latest patch release includes fixes to Silkworm fork handling behavior and improved connection settings for eos-evm-miner.
This release contains changes to EOS EVM Node, EOS EVM RPC, and tx-proxy.
Read on for more details.
Updates
Node fork handling behavior
Shut down EOS EVM Node when encountering error during block processing
Unexpected behavior was observed where Silkworm would continue to attempt to process transactions in later blocks when a previous block necessary to execute could not be found in the database, leading to a wrong nonce error. The behavior has been updated to shut down the node in these scenarios instead of continuing to attempt to process transactions.
Always ensure block is added to database regardless of header existing in cache
It was discovered that the fork switching logic in Silkworm does not clear the block header from the cache, which can lead to an edge case where a node recovering from a fork cannot add a block to the database when a block header is already previously stored. An update has been made to add blocks to the database even when a block header may be stored in the cache.
Miner connection settings
Use keepalive in tx-proxy to resolve http connection limit for miner
During a high traffic event the http connection limit in tx-proxy was being hit for EOS EVM Miner. The keepalive setting has been implemented for this connection to enable long connections from tx_proxy to miner as a workaround.
Building, compatibility, and upgrading
Building
The README in the eos-evm-node repository contains instructions on how to build EOS EVM Node and EOS EVM RPC.
Compatibility and upgrading
EOS EVM Node and EOS EVM RPC can be upgraded from the prior version by replacing the binaries.
Further details on changes since last release
Contributors
Special thanks to the contributors that submitted patches for this release:
Full list of changes since last release
PRs
