0.23.0 - 2026-09-29

What's Changed

Fixes

• fix(restore): circuit breaker config + stop global restore stalls (#197) (#199)

Dependencies

• deps: bump p256 from 0.13.2 to 0.14.0 (#128)
• deps: bump the rust-minor group across 1 directory with 21 updates (#200)
• ci: bump docker/login-action from 4.1.0 to 4.6.0 (#122)
• ci: bump dtolnay/rust-toolchain (#123)
• ci: bump docker/build-push-action from 7.1.0 to 7.4.0 (#124)
• ci: bump codecov/codecov-action from 6.0.1 to 7.1.1 (#126)
• ci: bump actions/checkout from 6.0.2 to 7.0.0 (#125)

Other

• ci: bump docker/login-action from 4.1.0 to 4.6.0 (#122)
• ci: bump dtolnay/rust-toolchain (#123)
• ci: bump docker/build-push-action from 7.1.0 to 7.4.0 (#124)
• ci: bump codecov/codecov-action from 6.0.1 to 7.1.1 (#126)
• ci: bump actions/checkout from 6.0.2 to 7.0.0 (#125)

Full Changelog: v0.22.0...v0.23.0

---

Release Notes

Added

• backup.circuit_breaker / restore.circuit_breaker YAML settings
  (enabled, failure_threshold, reset_timeout_ms, success_threshold)
  so operators can tune or disable the Kafka circuit breaker for large
  restores. Defaults match the previous hardcoded Kafka thresholds
  (failure=5, reset=30s, success=2). The breaker remains advisory:
  engines record successes/failures and log state transitions but do not
  block requests when Open — transient failures are retried by
  PartitionLeaderRouter. (#197)
• CircuitBreaker::disabled / CircuitBreaker::from_settings and
  RestoreEngine::kafka_circuit_state() for observability and tests.
• OffsetMapping::add_detailed_batch for batched per-record offset mapping.

Fixed

• Restore no longer takes the shared offset_mapping mutex once per record;
  range updates and detailed pairs are applied per segment / produce batch,
  cutting lock contention under max_concurrent_partitions > 1.
  (#197)
• PartitionLeaderRouter connection-error retries now evict only the pool
  of the broker the failed request was sent to (captured at dispatch, so a
  concurrent leader change cannot make the retry drop the new leader's
  healthy pool) instead of clearing every broker's pool; NOT_LEADER retries
  (produce, fetch, get_offsets) refresh metadata without dropping any pool,
  since the broker that answered is healthy and usually still leads other
  partitions of the same run. A blip on one leader therefore no longer forces
  unrelated partitions to rebuild TCP/TLS/SASL connections. Concurrent pool
  rebuilds keep an already-installed full pool instead of racing insert.
  (#197)
• OffsetMapping detailed mappings now stay sorted by source offset even
  when a record filter maps a segment's dropped records after its survivors,
  so lookup_target_offset always takes the exact binary-search path.

Changed

• Breaking (library API): BackupOptions and RestoreOptions gain
  circuit_breaker: CircuitBreakerSettings; CircuitBreakerConfig gains
  enabled: bool (default true). Consumers that build these structs by
  literal must add the fields — ..Default::default() where available.
• Dependencies: p256 0.14 (RustCrypto 0.14 generation — evidence signing
  and verification unchanged) and the routine minor/patch refresh of
  futures, serde, rustls, object_store, hyper, regex, uuid,
  tempfile and friends (#128,
  #200).

Install kafka-backup-cli 0.23.0

Install prebuilt binaries via shell script

curl --proto '=https' --tlsv1.2 -LsSf https://github.com/osodevops/kafka-backup/releases/download/v0.23.0/kafka-backup-cli-installer.sh | sh

Install prebuilt binaries via powershell script

powershell -ExecutionPolicy Bypass -c "irm https://github.com/osodevops/kafka-backup/releases/download/v0.23.0/kafka-backup-cli-installer.ps1 | iex"

Install prebuilt binaries via Homebrew

brew install osodevops/tap/kafka-backup

Download kafka-backup-cli 0.23.0

File	Platform	Checksum
kafka-backup-cli-aarch64-apple-darwin.tar.xz	Apple Silicon macOS	checksum
kafka-backup-cli-x86_64-apple-darwin.tar.xz	Intel macOS	checksum
kafka-backup-cli-x86_64-pc-windows-msvc.zip	x64 Windows	checksum
kafka-backup-cli-x86_64-unknown-linux-gnu.tar.xz	x64 Linux	checksum