CBSA Practice Questions

This is a correct definition of probabilistic finality. The transaction is considered 'final' after a certain number of confirmations (e.g., 6 for Bitcoin), but there is always a non-zero, though exponentially decreasing, probability of a chain reorganization that could reverse it.

This is a correct definition of absolute or deterministic finality. In systems like pBFT, once a supermajority of validators agrees on a block, it is considered final and cannot be reverted, short of a catastrophic failure of the consensus assumptions.

This statement is false. While difficult and expensive to execute against large networks like Bitcoin, 51% attacks are a very real threat and have been successfully carried out against several smaller but significant Proof of Work blockchains, including Ethereum Classic (ETC), Bitcoin Gold (BTG), and Verge (XVG). These attacks allowed the perpetrators to double-spend coins and reorganize parts of the chain.

Hashgraph, used by Hedera, is a consensus algorithm that utilizes a Directed Acyclic Graph (DAG) structure. Unlike traditional blockchains that have a single chain of blocks, DAGs allow transactions to be added in parallel, referencing multiple previous transactions. This structure, combined with a gossip protocol and virtual voting, enables extremely high throughput and fast finality, making it suitable for high-performance applications like gaming.

The most effective and standard pattern to prevent flash loan governance attacks is to take a snapshot of token balances at a specific block height, typically the block when the proposal was created. Voting power is then determined by the balance at that past block, not the current block. Since a flash loan is borrowed and returned within a single transaction (and thus a single block), the attacker's balance at the snapshot block would be their original, pre-loan amount, rendering the attack ineffective.

A Merkle Tree is a data structure that efficiently summarizes and verifies the integrity of large sets of data. In a Bitcoin block, all transaction hashes are repeatedly hashed together until a single hash, the Merkle Root, remains. This root is included in the block header. This allows for quick verification that a specific transaction is part of a block without downloading the entire block, a feature used by Simplified Payment Verification (SPV) clients.

Practical Byzantine Fault Tolerance (pBFT) is the ideal choice for this scenario. It is designed for permissioned networks, provides immediate transaction finality, offers low latency, and is resilient to f = (n-1)/3 faulty nodes, which matches the requirement of tolerating up to one-third malicious or offline nodes. PoW has probabilistic finality and high latency. PoS is generally for public networks and also has probabilistic finality in many implementations. RAFT is not Byzantine fault-tolerant and cannot handle malicious actors.

Private Data Collections (PDCs) are specifically designed for this use case. They allow a subset of organizations on a channel to endorse, commit, and query private data without having to create a separate channel. The actual private data is stored peer-to-peer in a separate state database, while a hash of the data is written to the channel's public ledger, providing the required auditability for all members without exposing the sensitive PII.

IPFS provides content-addressed, decentralized storage. Files are identified by a unique hash (CID), which can be stored on-chain efficiently.

This is the correct pattern. The large file is stored off-chain in a decentralized system like IPFS, and the small, immutable CID is stored on-chain as a verifiable link to the data.

This statement is true. The 'nothing at stake' problem is a fundamental challenge in early PoS designs. Unlike Proof of Work, where miners must expend significant computational resources (and thus cost) to mine on a fork, PoS validators can sign blocks on multiple competing chains with negligible cost. This can hinder consensus finality. Modern PoS systems address this with 'slashing' mechanisms, which penalize validators by taking away some of their staked tokens if they are caught signing conflicting blocks.

A decentralized oracle network (DON), such as the one used by Chainlink, is the industry-standard solution. It mitigates the risk of a single point of failure by using multiple independent oracle nodes and multiple premium data sources. The nodes fetch data, and their responses are aggregated on-chain to form a single, highly reliable data point, protecting the DeFi protocol from faulty or malicious data providers.