KIP-856 — KRaft Disk Failure Recovery
Discarded KRaft
Proposes a recovery path for KRaft controllers when the `metadata.log.dir` disk fails, by allowing the controller to be restarted with a fresh metadata log and re-synced from quorum peers. Currently a disk failure on the metadata log directory is unrecoverable without manual intervention or cluster recreation.
Details
| Author | José Armando García Sancio |
| Status | Discarded |
| JIRA | KAFKA-14113 |
| Wiki | View on Apache Wiki |
| Created | 2022-07-26 |
| Last Modified | 2024-03-29 |
Explore how this KIP affects the Kafka protocol in the Protocol Explorer, or see the full KIP database.