Transaction Monitoring of Smart Contracts

TL;DR

This paper explores the challenges of implementing transaction monitoring in smart contracts on blockchains and proposes mechanisms to enable it, addressing a key reliability concern.

Contribution

It demonstrates the limitations of transaction monitoring in most blockchains and proposes new execution mechanisms to enable it.

Findings

Transaction monitoring is generally not feasible in most existing blockchains.

Proposed execution mechanisms can enable effective transaction monitoring.

Enhances smart contract reliability through improved runtime verification.

Abstract

Blockchains are modern distributed systems that provide decentralized financial capabilities with trustable guarantees. Smart contracts are programs written in specialized programming languages running on a blockchain and govern how tokens and cryptocurrency are sent and received. Smart contracts can invoke other contracts during the execution of transactions initiated by external users. Once deployed, smart contracts cannot be modified and their pitfalls can cause malfunctions and losses, for example by attacks from malicious users. Runtime verification is a very appealing technique to improve the reliability of smart contracts. One approach consists of specifying undesired executions (never claims) and detecting violations of the specification on the fly. This can be done by extending smart contracts with additional instructions corresponding to monitor specified properties, resulting in an onchain monitoring approach. In this paper, we study transaction monitoring that consists of detecting violations of complete transaction executions and not of individual operations within transactions. Our main contributions are to show that transaction monitoring is not possible in most blockchains and propose different execution mechanisms that would enable transaction monitoring.