MegaETH decides to build its own RPC stack, expecting downtime during testing

On March 8th, MegaETH co-founder Lei Yang posted on social media that yesterday's RPC outage incident exposed an interesting edge case. The call to debug the RPC method caused the server to allocate a large amount of memory for the very long string, which eventually led to memory exhaustion. The MegaETH team decided to build their own RPC stack in order to apply a number of optimizations on the RPC stack to improve performance. From a performance perspective, it worked well, but at the cost of the team handling the operation and maintenance of the infrastructure itself. It is expected that there will be more outages in the future, and more post-mortem analysis reports will be released, hoping that these reports will spark interesting discussions. It encourages everyone to provide more edge cases for testing now, rather than having problems in a few months.