Releases: tis24dev/proxsave
Releases · tis24dev/proxsave
Release list
v0.37.0
Release Notes
-
New Features
- Added configuration integrity checks for duplicate, missing, unknown, and legacy variables, with actionable findings and verdicts.
- Improved personal-script diagnostics with clearer configuration explanations and hard-link advisories.
- Restore operations now report refused or indeterminate storage settings, overridden guest locks, and interrupted guest registrations more clearly.
-
Bug Fixes
- Personal-script launches now return when paths are blocked.
- Dashboard relaunches preserve the relaunched process’s exit status.
- Release notes are marked seen when closed by keyboard actions.
-
Documentation
- Updated configuration, daemon-status, installation, and troubleshooting guidance.
- Clarified email notification terminology by referring to delivery “modes.”
Changelog
v0.36.0
Release Notes
- New Features
- Added safer post-upgrade finalization using the newly installed version.
- Added clearer daemon-status reporting for unreadable configuration files, including
UNKNOWNstates and reasons. - Added personal-script hardlink protection advisories.
- Bug Fixes
- Prevented stderr warnings from corrupting PVE, PBS, rclone, and email-recipient data parsing.
- Improved guest and storage configuration fallback handling.
- Cancellations now stop staged configuration changes safely.
- Automatic upgrades no longer pause for the What's New screen.
- Rejected
--dry-runwhen used with upgrade operations.
- Documentation
- Updated CLI, daemon-status, security, and What's New documentation.
Changelog
v0.35.0
Release Notes
- New Features
- Daemon status compares running and current personal-script configurations, including synchronization, drift, warnings, and refusal reasons.
- Runtime details include configuration path, startup time, and availability status.
- Personal scripts under trusted user-owned parent directories can run with a
READY WITH WARNINGadvisory. - PVE cluster database snapshots can now use SQLite successfully.
- Bug Fixes
- Daemon runtime state is cleaned up reliably during shutdown.
- Personal scripts remain bound to validated files during execution.
- Documentation
- Updated guidance for script trust, runtime diagnostics, and daemon status.
Changelog
v0.34.0
Release Notes
- New Features
- Prevented multiple backup daemon instances from running simultaneously.
- Improved Proxmox VE cluster restore handling for guests, storage, datacenter settings, and scheduled jobs.
- Added more reliable Proxmox configuration database snapshots.
- Dashboard upgrades now relaunch the updated application automatically.
- Daemon status now reports personal script readiness and diagnostic details without executing scripts.
- Notification failures now produce a warning exit status for monitoring.
- Bug Fixes
- Corrected temporary-file detection, cloud log deletion, mount resolution, and storage completion reporting.
- Protected binary and non-plain-text files during text cleanup.
- Improved backup failure and skipped-operation logging.
- Added safer validation for cluster restore writes.
- Console log-level filtering now preserves warnings and errors in logs and counters.
- Documentation
- Updated CLI logging, notification, restore, daemon, and troubleshooting documentation.
Changelog
v0.33.0
Release Notes
-
New Features
- Added optional daemon-only pre-run and post-run scripts for supervised backups.
- Added trusted script-path validation with warnings for rejected paths.
-
Bug Fixes
- Improved storage and cancellation error messages with clearer causes and context.
- Secondary-storage listings now report unreadable archives and missing locations more accurately.
- Prevented failures when handling invalid storage errors.
-
Improvements
- Standardized log severity formatting across the application.
- Log categories now sort case-insensitively.
- Expanded configuration, security, daemon, and troubleshooting documentation.
Changelog
v0.32.0
Release Notes
- New Features
- Upgrades preserve the recorded scheduler mode and install the daemon only when no scheduler has been configured.
- Added reliable external upgrade support through
install.sh --upgrade. - Daemon and cron transitions now preserve schedules and report removed, retained, or duplicate entries.
- Bug Fixes
- Improved detection of wrapped and environment-prefixed cron commands.
- Healthcheck warnings now explain unsupported cron configurations.
- Cron scanning no longer hangs on unresponsive files or named pipes.
- Documentation
- Updated upgrade, daemon, cron, healthcheck, and installation guidance.
- Removed the deprecated daemon opt-out setting.
Changelog
v0.31.0
Release Notes
-
Bug Fixes
- Improved backup retention across hostname changes while protecting archives belonging to other hosts.
- Added safer handling for legacy and unattributable archives.
- Restore operations now validate archive ownership using the run’s resolved hostname.
- Strengthened restore path checks to block unsafe archive entries.
- Prevented command and diagnostic operations from hanging indefinitely.
- Corrected backup counts and retention reporting to reflect host-owned archives accurately.
-
Documentation
- Expanded cloud retention, troubleshooting, and release guidance.
- Release-note checks now apply consistently to prereleases.
Changelog
v0.30.0
Release Notes
First stable of the 0.30 line. Includes everything from v0.30.0-beta1 to v0.30.0-beta11.
-
New Features
- Rebuilt graphical interface: running
proxsaveopens an interactive dashboard, and backup, install, upgrade and support stream inside a contained scrollable panel. The legacy wizard is gone. - Resident daemon with its own scheduler and hang watchdog, replacing cron (reversible), plus
--daemon-status, binary-alignment checks and stuck-backup recovery. - Backup monitoring via per-channel Healthchecks sensors, centralized or self mode, with credential provisioning and recovery and a one-time portal sign-in link.
- In-app update screen and
--upgrade --localfile,upgrade-beta.shfor prereleases, and a What's New screen (--show-whatsnew) shown once after an interactive upgrade. HOST_BACKUP_MODEto back up a Proxmox host from an HA-LXC appliance over a read-only mount.- Richer backup recap and summaries: warnings, missing items, log path, Telegram Server ID, Healthchecks link, PVE collection summary.
- Full restore adds safety checks, confirmation and a safety backup; retention protects backups belonging to other hosts.
- Rebuilt graphical interface: running
-
Bug Fixes
- Around 200 fixes across restore, the graphical UI, install, the daemon, the dashboard, backup, storage, decrypt and security.
- Notification failures are scoped to a dedicated notify-error severity, with real Healthchecks transmission status and less duplicate logging.
- Telegram relay credentials are re-provisioned when rejected or parked, with bounded backoff.
- PBS notification restore reporting distinguishes empty, failed, partially applied and incomplete configurations.
-
Security
- Trojan-source build guard, operator-file reads confined through the OS root-directory API, and restore-safety fixes for symlinks, hardlinks, dedup scoping and atomic writes.
- Support backups use a secret maintainer address, shared consent prompts and stricter archive permissions.
- Refreshed security tooling, workflow pins, minimum Go runtime, and the release credential is no longer persisted in
.git/config.
-
Documentation
- New healthchecks and monitoring guide, plus expanded scheduling, storage, recovery, encryption, install, notification, restore and exit-code references.
-
Upgrade Notes
- Scheduling migrates from cron to the daemon automatically; it is reversible from the dashboard.
- PBS datastore file scanning is now off by default. Set
PXAR_SCAN_ENABLE=trueinbackup.envto keep it. - Open Healthchecks in the dashboard for the monitoring portal address and sign-in details.
Changelog
v0.30.0-beta11
Release Notes
-
New Features
- Improved backup summaries with compact and detailed views, clearer warnings, and artifact paths.
- Added recovery handling for stuck backup processes, including degraded health reporting and automatic daemon restart.
- Added shared consent prompts for support requests.
- Improved installation flows, validation, and setup feedback.
- Added safer restore behavior that excludes protected cluster database files.
-
Bug Fixes
- Cleanup guard operations now accurately report when protections remain.
- Backup scanning is disabled by default.
- Improved status-message sanitization and sensitive archive permissions.
-
Documentation
- Expanded encryption, daemon recovery, healthcheck, restore, upgrade, and exit-code guidance.
Changelog
v0.30.0-beta10
Release Notes
-
New Features
- Healthchecks now displays a secure one-time sign-in link or sanitized portal address and login details.
- Installation and upgrades can adopt an existing daily schedule when no time is configured.
- Full restore includes safety checks, confirmation, and a safety backup.
- Retention protects backups belonging to other hosts.
-
Bug Fixes
- Improved portal validation, scheduler handling, restore safety, and retention accuracy.
-
Documentation
- Expanded guidance for scheduling, monitoring, storage, recovery, encryption, installation, security, notifications, and exit codes.