- New internal/fs package for testable filesystem operations
- In-memory filesystem support for unit testing without disk I/O
- Swappable global FS: SetFS(afero.NewMemMapFs())
- Wrapper functions: ReadFile, WriteFile, Mkdir, Walk, Glob, etc.
- Testing helpers: WithMemFs(), SetupTestDir()
- Comprehensive test suite demonstrating usage
- Upgraded afero from v1.10.0 to v1.15.0
- Windows-compatible colors via native console API
- Color helper functions: Success(), Error(), Warning(), Info()
- Text styling: Header(), Dim(), Bold(), Green(), Red(), Yellow(), Cyan()
- Logger CleanFormatter uses fatih/color instead of raw ANSI
- All progress indicators use colored [OK]/[FAIL] status
- Automatic color detection for non-TTY environments
- Visual progress bars for cloud uploads/downloads
- Byte transfer display, speed, ETA prediction
- Color-coded Unicode block progress
- Checksum verification with progress bar for large files
- Spinner for indeterminate operations (unknown size)
- New types: NewSchollzBar(), NewSchollzBarItems(), NewSchollzSpinner()
- Progress Writer() method for io.Copy integration
- Use hashicorp/go-multierror for cluster restore error collection
- Shows ALL failed databases with full error context (not just count)
- Bullet-pointed output for readability
- Thread-safe error aggregation with dedicated mutex
- Error wrapping with %w for proper error chain preservation
- Added gopsutil/v3 for cross-platform system metrics
* Works on Linux, macOS, Windows, BSD
* Memory detection no longer requires /proc parsing
- Added go-humanize for readable output
* humanize.Bytes() for memory sizes
* humanize.Comma() for large numbers
- Improved preflight display with memory usage percentage
- Linux kernel checks (shmmax/shmall) still use /proc for accuracy
- Linux system checks (read-only from /proc, no auth needed):
* shmmax, shmall kernel limits
* Available RAM check
- PostgreSQL auto-tuning:
* max_locks_per_transaction scaled by BLOB count
* maintenance_work_mem boosted to 2GB for faster indexes
* All settings auto-reset after restore (even on failure)
- Archive analysis:
* Count BLOBs per database (pg_restore -l or zgrep)
* Scale lock boost: 2048 (default) → 4096/8192/16384 based on count
- Nice TUI preflight summary display with ✓/⚠ indicators
- Automatically boost max_locks_per_transaction to 2048 before restore
- Uses ALTER SYSTEM + pg_reload_conf() - no restart needed
- Automatically resets to original value after restore (even on failure)
- Prevents 'out of shared memory' OOM on BLOB-heavy SQL format dumps
- Works transparently - no user intervention required
- Auto-detect large objects in pg_restore dumps
- Split restore into pre-data, data, post-data phases
- Each phase commits and releases locks before next
- Prevents 'out of shared memory' / max_locks_per_transaction errors
- Updated error hints with better guidance for lock exhaustion
- Increase timeout from 60 to 180 minutes for very large archives
- Use streaming pipes instead of buffering entire tar listing
- Only mark as corrupted for clear corruption signals (unexpected EOF, invalid gzip)
- Prevents false CORRUPTED errors on valid large archives
- CheckDiskSpace now uses GetEffectiveWorkDir() instead of BackupDir
- Dynamic timeout calculation based on file size:
- diagnoseClusterArchive: 5 + (GB/3) min, max 60 min
- verifyWithPgRestore: 5 + (GB/5) min, max 30 min
- DiagnoseClusterDumps: 10 + (GB/3) min, max 120 min
- TUI safety checks: 10 + (GB/5) min, max 120 min
- Timeout vs corruption differentiation (no false CORRUPTED on timeout)
- Streaming tar listing to avoid OOM on large archives
For 119GB archives: ~45 min timeout instead of 5 min false-positive
- verifyArchiveCmd now uses restore.Safety and restore.Diagnoser
- Same validation logic in backup manager verify and restore safety checks
- No more discrepancy between verify showing valid and restore failing
- Service templates now use WorkingDirectory for config loading
- Config is read from .dbbackup.conf in /var/lib/dbbackup
- Updated SYSTEMD.md documentation to match actual CLI
- Removed non-existent --config flag from ExecStart
- Block cluster restore if existing databases found and cleanup not enabled
- User must press 'c' to enable 'Clean All First' before proceeding
- Prevents accidental data conflicts during disaster recovery
- Bug #24: Missing safety gate for cluster restore
CRITICAL FIXES:
- Encryption detection false positive (IsBackupEncrypted returned true for ALL files)
- 12 cmd.Wait() deadlocks fixed with channel-based context handling
- TUI timeout bugs: 60s->10min for safety checks, 15s->60s for DB listing
- diagnose.go timeouts: 60s->5min for tar/pg_restore operations
- Panic recovery added to parallel backup/restore goroutines
- Variable shadowing fix in restore/engine.go
These bugs caused pg_dump backups to fail through TUI for months.
This is a critical bugfix release addressing multiple hardcoded temporary directory paths
that prevented proper use of the WorkDir configuration option.
PROBLEM:
Users configuring WorkDir (e.g., /u01/dba/tmp) for systems with small root filesystems
still experienced failures because critical operations hardcoded /tmp instead of respecting
the configured WorkDir. This made the WorkDir option essentially non-functional.
FIXED LOCATIONS:
1. internal/restore/engine.go:632 - CRITICAL: Used BackupDir instead of WorkDir for extraction
2. cmd/restore.go:354,834 - CLI restore/diagnose commands ignored WorkDir
3. cmd/migrate.go:208,347 - Migration commands hardcoded /tmp
4. internal/migrate/engine.go:120 - Migration engine ignored WorkDir
5. internal/config/config.go:224 - SwapFilePath hardcoded /tmp
6. internal/config/config.go:519 - Backup directory fallback hardcoded /tmp
7. internal/tui/restore_exec.go:161 - Debug logs hardcoded /tmp
8. internal/tui/settings.go:805 - Directory browser default hardcoded /tmp
9. internal/tui/restore_preview.go:474 - Display message hardcoded /tmp
NEW FEATURES:
- Added Config.GetEffectiveWorkDir() helper method
- WorkDir now respects WORK_DIR environment variable
- All temp operations now consistently use configured WorkDir with /tmp fallback
IMPACT:
- Restores on systems with small root disks now work properly with WorkDir configured
- Admins can control disk space usage for all temporary operations
- Debug logs, extraction dirs, swap files all respect WorkDir setting
Version: 3.42.1 (Critical Fix Release)
- Remove invalid --config flag from exporter service template
- Change ReadOnlyPaths to ReadWritePaths for catalog access
- Add copyBinary() to install binary to /usr/local/bin (ProtectHome compat)
- Fix exporter status detection using direct systemctl check
- Add os/exec import for status check
Systemd Integration:
- New 'dbbackup install' command creates service/timer units
- Supports single-database and cluster backup modes
- Automatic dbbackup user/group creation with proper permissions
- Hardened service units with security features
- Template units with configurable OnCalendar schedules
- 'dbbackup uninstall' for clean removal
Prometheus Metrics:
- 'dbbackup metrics export' for textfile collector format
- 'dbbackup metrics serve' runs HTTP exporter on port 9399
- Metrics: last_success_timestamp, rpo_seconds, backup_total, etc.
- Integration with node_exporter textfile collector
- --with-metrics flag during install
Technical:
- Systemd templates embedded with //go:embed
- Service units include ReadWritePaths, OOMScoreAdjust
- Metrics exporter caches with 30s TTL
- Graceful shutdown on SIGTERM
PROBLEM: Users could not interrupt backup or restore operations through
the TUI interface. Pressing Ctrl+C or ESC did nothing during execution.
ROOT CAUSE:
- BackupExecutionModel ignored ALL key presses while running (only handled when done)
- RestoreExecutionModel returned tea.Quit but didn't cancel the context
- The operation goroutine kept running in the background with its own context
FIX:
- Added cancel context.CancelFunc to both execution models
- Create child context with WithCancel in New*Execution constructors
- Handle ctrl+c and esc during execution to call cancel()
- Show 'Cancelling...' status while waiting for graceful shutdown
- Show cancel hint in View: 'Press Ctrl+C or ESC to cancel'
The fix works because:
- exec.CommandContext(ctx) will SIGKILL the subprocess when ctx is cancelled
- pg_dump, pg_restore, psql, mysql all get terminated properly
- User sees immediate feedback that cancellation is in progress
- Add identifier validation for database names in PostgreSQL and MySQL
- validateIdentifier() rejects names with invalid characters
- quoteIdentifier() safely quotes identifiers with proper escaping
- Max length: 63 chars (PostgreSQL), 64 chars (MySQL)
- Only allows alphanumeric + underscores, must start with letter/underscore
- Fix data race in notification manager
- Multiple goroutines were appending to shared error slice
- Added errMu sync.Mutex to protect concurrent error collection
- Security improvements prevent:
- SQL injection via malicious database names
- CREATE DATABASE `foo`; DROP DATABASE production; --`
- Race conditions causing lost or corrupted error data
- Validate SQL dump files BEFORE attempting restore
- Detect unterminated COPY blocks that cause 'syntax error' failures
- Cluster restore now pre-validates ALL dumps upfront (fail-fast)
- Saves hours of wasted restore time on corrupted backups
The truncated resydb.sql.gz was causing 49min restore attempts
that failed with 2.6M errors. Now fails immediately with clear
error message showing which table's COPY block was truncated.
- Added WorkDir to Config for custom temp directory
- TUI Settings: new 'Work Directory' option to set alternative temp location
- Restore Preview: press 'w' to toggle work directory (uses backup dir as default)
- Diagnose View: now uses configured WorkDir for cluster extraction
- Config persistence: WorkDir saved to .dbbackup.conf
This fixes diagnosis/restore failures when /tmp is too small for large archives.
Use cases: servers with limited /tmp, 70GB+ archives needing 280GB+ extraction space.
- Better error messages when tar extraction fails
- Detect truncated/corrupted archives without full extraction
- Show archive contents even when extraction fails
- Provide helpful hints for disk space and corruption issues
- Exit status 2 from tar now shows detailed diagnostics
- Added 'Diagnose Backup File' as menu option in TUI
- Archive browser now supports 'diagnose' mode
- Allows users to run deep diagnosis on backups before restore
- Helps identify truncation/corruption issues in large backups
Features:
- restore diagnose command for backup file analysis
- Deep COPY block verification for truncated dump detection
- PGDMP signature and gzip integrity validation
- Detailed error reports with --save-debug-log flag
- Ring buffer stderr capture (prevents OOM on 2M+ errors)
- Error classification with actionable recommendations
TUI Enhancements:
- Automatic dump validity safety check before restore
- Press 'd' in archive browser to diagnose backups
- Press 'd' in restore preview for debug log toggle
- Debug logs saved to /tmp on failure when enabled
Documentation:
- Updated README with diagnose command and examples
- Updated CHANGELOG with full feature list
- Updated restore preview screenshots
Fixes Windows, OpenBSD, and NetBSD builds by extracting
EstimateBackupSize from disk_check.go (which has build tags
excluding those platforms) to a new estimate.go file.
Why wrap external tools when you can BE the tool?
New physical backup engines:
• MySQL Clone Plugin - native 8.0.17+ physical backup
• Filesystem Snapshots - LVM/ZFS/Btrfs orchestration
• Binlog Streaming - continuous backup with seconds RPO
• Parallel Cloud Upload - stream directly to S3, skip local disk
Smart engine selection automatically picks the optimal strategy based on:
- MySQL version and edition
- Available filesystem features
- Database size
- Cloud connectivity
Zero external dependencies. Single binary. Enterprise capabilities.
Commercial backup vendors: we need to talk.
- Add --dry-run/-n flag for backup commands with comprehensive preflight checks
- Database connectivity validation
- Required tools availability check
- Storage target and permissions verification
- Backup size estimation
- Encryption and cloud storage configuration validation
- Implement GFS (Grandfather-Father-Son) retention policies
- Daily/Weekly/Monthly/Yearly tier classification
- Configurable retention counts per tier
- Custom weekly day and monthly day settings
- ISO week handling for proper week boundaries
- Add notification system with SMTP and webhook support
- SMTP email notifications with TLS/STARTTLS
- Webhook HTTP notifications with HMAC-SHA256 signing
- Slack-compatible webhook payload format
- Event types: backup/restore started/completed/failed, cleanup, verify, PITR
- Configurable severity levels and retry logic
- Update README.md with documentation for all new features
- Add 'migrate cluster' command for full cluster migration
- Add 'migrate single' command for single database migration
- Support PostgreSQL and MySQL database migration
- Staged migration: backup from source → restore to target
- Pre-flight checks validate connectivity before execution
- Dry-run mode by default (--confirm to execute)
- Support for --clean, --keep-backup, --exclude options
- Parallel backup/restore with configurable jobs
- Automatic cleanup of temporary backup files
- Fix format detection to read database_type from .meta.json metadata file
- Add ensureMySQLDatabaseExists() for MySQL/MariaDB database creation
- Route database creation to correct implementation based on db type
- Add TUI auto-forward in auto-confirm mode (no input required for debugging)
- All TUI components now exit automatically when --auto-confirm is set
- Fix status view to skip loading in auto-confirm mode
