journal-guided-resynchronization · research

Journal-guided resynchronization

Journal-guided resynchronization uses filesystem recovery records to target RAID consistency checks after a crash.

Denehy and colleagues' FAST 2005 prototype combined ext3 declared mode with a verify-read request to software RAID. Declarations recorded data-write destinations before modification; verify-read checked a stripe and repaired inconsistent parity. Ordinary read-modify-write can retain a pre-existing parity error. [1]

Relations

related
Write-intent bitmap, RAID write hole, Partial Parity Log

Advantages

  • Reuses cross-layer information to narrow recovery work.

Limitations

  • Historical prototype, not a current ext4 setting.
  • Requires coordinated filesystem and RAID changes, not merely an existing journal.

Citations

  1. Denehy, Arpaci-Dusseau and Arpaci-Dusseau. Journal-guided Resynchronization for Software RAID. FAST 2005, sections 4-5. (accessed 2026-10-25)

Further reading

  1. Hesela: RAID write-hole recovery mechanisms (accessed 2026-10-25)

JSON: /data/terms/journal-guided-resynchronization.json

Published by Hesela team · Updated

Report a correction · Editorial standards · Reuse guidance

Cite this page
Hesela team. Journal-guided resynchronization. Hesela. 2026-10-25. https://hesela.dev/terms/journal-guided-resynchronization/