Skip to content

Aerospike Backup Service 3.5.0 release notes

March 10, 2026  |  Download

New in this version

Aerospike Backup Service 3.5.0 enables concurrent full and incremental backups, adds safeguards to prevent restores during active backups, and introduces API-triggered incremental backups. This release also improves restore reliability by using only finished backups and adds storage connectivity checks at startup.

New features

  • Full and incremental backups can now run concurrently, allowing for shorter Recovery Point Objectives (RPO) by eliminating wait times between backup types.
  • Restores are now disallowed on the same cluster/namespace while an existing full or incremental backup is ongoing.
  • Added a new API endpoint to trigger incremental backups on demand.
  • Restore operations now use only successfully completed backups, preventing potential data inconsistency from in-progress backup files.
  • The last_backup_time Prometheus metric now includes a type label to distinguish between full and incremental backups.
  • Added validation checks for restore operations to catch configuration errors before the restore begins.
  • Backup runs now report an explicit “canceled” outcome with improved logging.
  • Added proactive storage connectivity validation during service startup with checks against configured S3, GCP, and Azure resources.
  • Streamlined the restore-by-timestamp API request format by allowing destination cluster and secret agent to default from the routine when omitted, and by inferring compression settings from backup metadata.
  • Added a min-part-size configuration option for local storage to control chunk size when writing backups to disk.

Improvements

  • Enhanced resilience to Aerospike cluster reconfiguration events.
  • Made restore job operations thread-safe.
  • Made TLS-related validations optional for RestoreRequest validation, so missing TLS files no longer block config validation workflows.
  • Added detailed logging for restore operation results, including successful, canceled, failed-to-start (precondition), and failed outcomes.
  • Improved HTTP server lifecycle handling by surfacing startup failures and preserving graceful shutdown behavior.

Bug fixes

  • Fixed backup discovery for restore operations to correctly filter by namespace.
  • Resolved incorrect invalidation of backup routines under certain conditions.
  • Passwords read from files are now properly trimmed of whitespace.
  • Added minimal delay for ad-hoc backups to prevent race conditions in rapid successive requests.
  • Ensured routine configurations remain immutable during execution.

For additional details about fixes, improvements, and maintenance in this release, see the release page on GitHub.

Feedback

Was this page helpful?

What type of feedback are you giving?

What would you like us to know?

+Capture screenshot

Can we reach out to you?