fix(cluster): improve shard reconstruction validation and start background healing service
This commit is contained in:
@@ -1,5 +1,14 @@
|
||||
# Changelog
|
||||
|
||||
## 2026-03-21 - 6.3.1 - fix(cluster)
|
||||
improve shard reconstruction validation and start background healing service
|
||||
|
||||
- use the erasure read quorum when reconstructing chunks instead of assuming data shard count
|
||||
- verify reconstructed shards before writing healed data back to disk
|
||||
- start the healing service during server initialization with shared local shard stores
|
||||
- simplify QUIC request handling by decoding the full request buffer including trailing shard data
|
||||
- clean up unused variables and imports across cluster modules
|
||||
|
||||
## 2026-03-21 - 6.3.0 - feat(readme)
|
||||
document distributed cluster mode, erasure coding, and QUIC-based architecture
|
||||
|
||||
|
||||
Reference in New Issue
Block a user