Raft log replication, durable state and liveness
Research context
research method · established · As of 2026-10-10
Raft uses term-numbered leader election, majority voting and log matching to replicate a state machine under crash faults.
Key knowledge
Raft uses term-numbered leader election, majority voting and log matching to replicate a state machine under crash faults. A leader commits an entry from its current term by majority replication; earlier entries become committed through the rule, not simply by counting any old replicas. Votes, current term and log must survive crashes. A connected majority and favorable timing are needed for progress; the safety argument does not make Raft Byzantine tolerant.
Conditions and scope
- Use a fixed membership of 2f+1 crash-fault nodes for tolerating f crashes; reconfiguration needs its own overlap protocol.
- State-machine execution, client retry deduplication and durable persistence must preserve the protocol's assumptions.
Sources · 1
References
Sources retain their recorded scope. A source link alone does not establish verification of every statement.
Evidence and review
Author checked this English statement, displayed equation where present and stated hypotheses against these source locations: §§5–8: leader election, log matching, commit rule, durable state, membership and client linearizability. The check is bounded to the recorded claims; no complete proof reconstruction, software proof replay or universal literature completeness is claimed.
2026-10-10 · Author source-body/statement/equation/domain check, with graph associations scoped individually. Separate independent content/source/inventory review is required before branch closure.
Entry details
Permanent link to this entry ↗- Record identity
- research-info-raft-safety-liveness
- Revision
- research.1
- Structured reference record
- Postgraduate & research reference