Blog

IBM Informix backup integrated with ON-Bar. Hot backup, individual dbspaces, logical log streaming, log-precise PITR, Continuous Log Restore (CLR) HA support.

• Native ON-Bar — direct integration with the official Informix subsystem.
• dbspace-level backup — backup and restore per dbspace, full granularity.
• Logical log streaming — near-zero RPO.
• Log-precise PITR — precise restore.
• HDR and Enterprise Replication aware — back up from secondary, zero overhead on primary.

Production-ready in 30 days, at least 50% cheaper. Free trial, signed RPMs and DEBs, direct support from Heitor Faria. Replace your Veeam, Commvault or Bacula Enterprise license without breaking the nightly window.

Request a free 30-day trial →

🔴 COMMERCIAL DISCLAIMER

Bring your Bacula Enterprise, Veeam, Commvault or Netbackup contract or renewal proposal. We offer at least 50% discount with significantly more features.

Contact: heitor@opentechs.lat | +1 786 726-1749 | +55 61 98268-4220 (WhatsApp)

---

Table of Contents

1. Executive Summary
2. Business Case & TCO
3. Architecture Overview
4. Backup Engines Deep Dive
5. Replication Capabilities
6. Advanced Features
7. Security & Compliance
8. Observability
9. Installation Guide
10. Configuration Reference
11. Restore Procedures & Use Cases
12. Performance Reports
13. Compatibility Matrix
14. Sizing Guide
15. Bacularis Integration
16. Migration from TSM/ISP and ON-Bar PSM
17. Troubleshooting
18. Roadmap
19. Validation Evidence (E2E Results)

---

1. Executive Summary

The PodHeitor Informix Backup, Replication and Conversion Plugin for Bacula is a commercial-grade, enterprise-ready plugin for HCL Informix Dynamic Server (IDS) 12.10 through 15.0, built natively for PodHeitor Backup+. Written entirely in Rust, it delivers near-zero runtime overhead, memory safety, and parallelism-first design — all in a single signed binary package that installs in minutes.

The Gap It Fills

Bacula Enterprise has no dedicated Informix plugin. The Enterprise database catalog ships DB2, MSSQL, MySQL, Oracle, PostgreSQL, Sybase and Amazon RDS — but not Informix. Organizations running IDS today are forced to choose between expensive IBM Spectrum Protect (TSM), manual bpipe+ontape shell scripts with no PITR or chain safety, or HCL’s PSM creating double-I/O staging on disk.

PodHeitor Informix is the first Bacula-native, XBSA-aware, Rust-based Informix plugin in the market.

Key Differentiators

Capability	Hand-rolled scripts	TSM/PSM + Bacula	PodHeitor Informix
Cost	Free, heavy maintenance	Very expensive	One-time license, no per-DB fees
Native XBSA	No	TSM only	Yes — libpodheitor_xbsa.so
Parallelism	None	Limited	N stripes × M dbspaces
PITR	Manual	Yes	Automated continuous log shipping
Single-table restore	None	Manual archecker	Built-in mode=table_level
Cross-version conversion	Risky	No	mode=logical_dbexport first-class
HDR/RSS/SDS/ER bootstrap	Manual hours	Manual hours	One bconsole command
Instant recovery	No	No	IDS online in < 60 s via FUSE
Ransomware protection	No	No	ARD hook (IDS 14.10+)
Observability	None	TSM dashboard	Prometheus + OTLP
Bacula-native	Partial (bpipe)	Two-layer	Direct XBSA + streaming

Validation Status

E2E Lab (2026-04-29):  PASS=22  FAIL=0  SKIP=0  TOTAL=22
Unit tests:            PASS=80  FAIL=0

All 22 end-to-end scenarios active and passing on a 4-VM lab (primary IDS, HDR secondary, RSS secondary, snapshot host) with IDS 15.0.1.0.3 Developer Edition and PodHeitor Backup.

---

2. Business Case & TCO

2.1 Cost Comparison (Annual, 10 TB Informix environment)

Solution	License	Infrastructure	Operational	Total
IBM Spectrum Protect	~$80,000/yr	Separate TSM server	DBA + TSM admin FTE	~$140,000/yr
Bacula Enterprise + manual scripts	~$30,000/yr	None	DBA FTE (scripts)	~$70,000/yr
Veeam Enterprise (DB module)	~$25,000/yr	None	Admin FTE	~$55,000/yr
PodHeitor Informix + PodHeitor Backup	One-time fee	Existing Bacula infra	Minimal	< $10,000 total

Contact heitor@opentechs.lat for pricing. We offer at least 50% discount on any competing renewal.

2.2 Risk Reduction

Risk	Without PodHeitor	With PodHeitor
Backup chain corruption	Silent — detected at restore time	Detected immediately via chain_health_check + archecker
Ransomware encryption before backup	Backup captures encrypted data	ARD hook aborts backup on anomaly
Point-in-time recovery gap	Manual log archive management	Continuous log shipping daemon, sub-minute RPO
HDR secondary bootstrap failure	Hours of manual work + outage risk	One-command automated bootstrap
Cross-version upgrade risk	Manual dbexport/dbimport + downtime	mode=logical_dbexport streaming, tested in E2E
Bare-metal DR (total server loss)	Requires manual onconfig/sqlhosts reconstruction	RestoreObject captures all metadata automatically

2.3 ROI Calculation Example

A mid-sized financial institution running Informix 14.10 on 5 servers, with daily backups and 4-hour RTO requirement:

• TSM license + infrastructure: ~$60,000/yr
• One recovery incident at 4h × 3 DBAs: ~$3,600 incident cost
• PodHeitor migration: one-time cost + PodHeitor Backup (free)
• Payback period: < 3 months

---

3. Architecture Overview

3.1 Component Diagram

┌─────────────────────────────────────────────────────────────────────┐
│  Bacula Director 15.0.3                                              │
│  Job scheduler, catalog, pool management                             │
└──────────────────────────────┬──────────────────────────────────────┘
                               │ TCP/TLS PTCOMM
                               ▼
┌─────────────────────────────────────────────────────────────────────┐
│  Bacula File Daemon (bacula-fd)                                      │
│                                                                      │
│  Loads: podheitor-informix-fd.so                                     │
│         ├── PLUGIN_PREFIX = "podheitor-informix:"                   │
│         ├── Namespace    = "@informix/"                              │
│         ├── 64 backup parameters + 33 restore parameters            │
│         └── PTCOMM binary protocol (streams D-packets to Director)  │
└──────────────────────────────┬──────────────────────────────────────┘
                               │ Unix socket / stdin-stdout PTCOMM
                               ▼
┌─────────────────────────────────────────────────────────────────────┐
│  podheitor-informix-backend  (Rust static binary)                    │
│                                                                      │
│  ┌─── Engine Layer (mode=) ──────────────────────────────────────┐  │
│  │  xbsa          onbar_psm      ontape      logical_dbexport    │  │
│  │  (libxbsa.so)  (FIFO drain)   (STDIO)     (dbexport stream)   │  │
│  │  snapshot      replicate_hdr  replicate_rss  replicate_er     │  │
│  └──────────────────────────────────────────────────────────────┘  │
│  ┌─── XBSA Library (libpodheitor_xbsa.so) ──────────────────────┐  │
│  │  BSAInit / BSABeginTxn / BSACreateObject / BSASendData        │  │
│  │  BSAEndTxn / BSATerminate / BSAGetData / BSAQueryObject       │  │
│  └──────────────────────────────────────────────────────────────┘  │
│  ┌─── Logical-Log Shipper Daemon ────────────────────────────────┐  │
│  │  Polls sysmaster:syslogs every N seconds                       │  │
│  │  Ships closed logs via PTCOMM — sub-minute RPO                 │  │
│  └──────────────────────────────────────────────────────────────┘  │
│  ┌─── Observability ─────────────────────────────────────────────┐  │
│  │  Prometheus /metrics    OpenTelemetry OTLP traces              │  │
│  └──────────────────────────────────────────────────────────────┘  │
│  ┌─── Security ──────────────────────────────────────────────────┐  │
│  │  DBCRYPT capture/restore   ARD ransomware hook   Redaction    │  │
│  └──────────────────────────────────────────────────────────────┘  │
└──────────────────────────────┬──────────────────────────────────────┘
                               │ XBSA ABI / pipes / sockets / SQL
                               ▼
┌─────────────────────────────────────────────────────────────────────┐
│  HCL Informix Dynamic Server (12.10.FC12+ / 14.10 / 15.0)           │
│  + sysmaster (bar_action, bar_object, syslogs, sysshmvals)          │
│  + ON-Bar (onbar) + archecker + oncheck                             │
│  + HDR / RSS / SDS / ER replication topology                        │
│  + DBCRYPT keystore                                                  │
└─────────────────────────────────────────────────────────────────────┘

3.2 Data Flow — XBSA Mode (Preferred)

ON-Bar invoked by plugin
    │
    │ dlopen(libpodheitor_xbsa.so)
    ▼
libpodheitor_xbsa.so  ──Unix socket──►  podheitor-informix-backend
    │                                         │
    │ BSACreateObject(dbspace)                 │  compress (zstd/lz4)
    │ BSASendData(blocks)                      │  stripe N PTCOMM streams
    │                                         │  compute SHA-256
    │                                         │  write manifest sidecar
    ▼                                         ▼
  BSAEndTxn                           Bacula Storage Daemon
                                            (File, Tape, S3, Dedup)

3.3 Backup Volume Namespace

/@informix/<INSTANCE_UUID>/
├── __ph_manifest.json          # canonical job manifest (LSN, SHA256, dbspace list)
├── __ph_topology.json          # HDR/RSS/SDS/ER topology snapshot
├── __ph_dbcrypt.json           # encrypted keystore reference (no plaintext key)
├── __ph_layout.json            # dbspace → chunk path mapping
├── restoreobjects/
│   ├── onconfig
│   ├── sqlhosts
│   └── ixbar.<servernum>       # ON-Bar emergency boot file
├── dbspace/<name>/
│   ├── L0.<jobid>.s<N>.zst    # level-0 stripe N (ZSTD frame)
│   ├── L1.<jobid>.s<N>.zst
│   └── manifest.<jobid>.json  # per-dbspace LSN + SHA256
├── logs/
│   ├── log.<unique_id>.<lsn>.zst
│   └── continuous/<YYYY>/<MM>/<DD>/
└── logical/
    ├── dbexport.<db>.<jobid>.tar.zst
    └── schema.<db>.<jobid>.sql

---

4. Backup Engines Deep Dive

4.1 Mode xbsa — Native XBSA (Default on mode=auto)

The strategic differentiator. libpodheitor_xbsa.so implements the full IBM XBSA C ABI, built in Rust with native memory-safety. ON-Bar loads it via BAR_BSALIB_PATH, making Bacula a first-class Informix storage manager — no staging, no double-I/O, no FIFO juggling.

Advantages over all alternatives:

• ON-Bar’s native parallelism (BAR_MAX_BACKUP) maps 1:1 to PTCOMM streams
• Full access to ON-Bar’s catalog (sysmaster:bar_action, bar_object) for LSN capture
• Supports archecker post-backup verification through standard ON-Bar book-keeping
• Zero bytes ever written to the FD host’s local disk (working dir < 50 MB)

XBSA functions implemented: BSAInit, BSABeginTxn, BSACreateObject, BSASendData, BSAEndTxn, BSATerminate, BSAGetData, BSAQueryObject, BSAEndData, BSADeleteObject

4.2 Mode onbar_psm — PSM Fallback

ON-Bar writes to HCL Primary Storage Manager (PSM) targeted at a named pipe. Backend drains the pipe → PTCOMM. Used when organization policy forbids loading 3rd-party XBSA libraries.

4.3 Mode ontape — Legacy/Simple

ontape -s -t STDIO -L 0|1|2. Whole-server only; no dbspace granularity. For dev, small instances, or environments without ON-Bar configured. Validated in E2E T16.

4.4 Mode logical_dbexport — Cross-Version Conversion

dbexport <db> -ss -X -q streamed through an async tar-on-the-fly pipeline → PTCOMM. Output is a compressed tar containing schema SQL + per-table unload files. Restorable into any IDS version with dbimport. Primary use: major version migration (12.10 → 14.10 → 15.0). Validated in E2E T12.

4.5 Mode snapshot — LVM/ZFS Quiesce

1. onmode -c checkpoint
2. Put dbspaces into backup mode (BAR_BLOCK_BACKUP=1)
3. LVM lvcreate --snapshot / ZFS zfs snapshot
4. Release backup mode — quiesce window < 1 second
5. Stream snapshot blocks via PTCOMM (content-defined chunking for dedup)
6. Capture archived logical logs via onbar -b -l

Validated in E2E T13 using loop-backed LVM VG (ifxvg/ifxdata).

4.6 Backup Level Mapping

Bacula Level	ON-Bar Level	Plugin Behavior
Full	onbar -b -L 0	Whole-server level 0 baseline. Manifest captures BAR_VERSION, INSTANCE_UUID, full dbspace list, LSN.
Differential	onbar -b -L 1	Cumulative since L0. Auto-promotes to Full if chain_health_check detects broken chain.
Incremental	onbar -b -L 2 OR logical-log only	Default: level_incr=logical_log for sub-minute RPO. Ships only newly-closed log files.
VirtualFull	(synthetic)	Plugin merges L0+L1+L2 chain into new L0 stream consumed by Bacula consolidation.

---

5. Replication Capabilities

5.1 HDR (High-Availability Data Replication) Bootstrap

Problem: Setting up an HDR secondary today requires manual onmode -d primary, onmode -d secondary, custom network config, often hours of DBA work.

PodHeitor solution: Single restore job with mode=ag_seed ag_target_role=hdr_secondary:

1. Restores latest backup chain to target with WITH NORECOVERY equivalent
2. Adjusts DRAUTO/DRINTERVAL/DRTIMEOUT on target
3. Executes onmode -d primary <target_servernum> on primary
4. Executes onmode -d secondary <primary_servernum> on target
5. Validates secondary reaches On-Line (Sec) state

Validated in E2E T10 and T14.

5.2 RSS (Remote Standalone Secondary) Bootstrap

Same flow with ag_target_role=rss. Uses onmode -d add RSS <name> on primary + onmode -d RSS <primary_name> on target. Validated in E2E T11.

5.3 SDS (Shared Disk Secondary) Attach

Validates shared-disk topology, attaches new SDS via onmode -d set SDS PRIMARY <primary_name>.

5.4 Enterprise Replication (ER) Topology Snapshot/Restore

__ph_topology.json captures the full ER mesh: servers, replicates, grids, subscriptions. On restore, cdr define server + cdr define replicate + cdr start replicate reconstitutes the topology. Critical for DR when the entire ER mesh must be rebuilt.

5.5 Backup from Secondary (Offload Primary)

Plugin = "podheitor-informix: prefer_secondary_for_backup=yes hdr_role_check=secondary"

Plugin detects the HDR pair, connects to the secondary, runs ON-Bar there. Primary CPU/I/O completely unaffected during backup window.

---

6. Advanced Features

6.1 Continuous Logical-Log Shipping (Sub-minute RPO)

When continuous_log=yes, the backend spawns a daemon thread that:

• Polls sysmaster:syslogs every continuous_log_interval seconds (default: 60s, min: 5s)
• Detects newly-closed logical log files (status U → B)
• Ships each log atomically via a PTCOMM micro-job with SHA-256 integrity check
• Detects and alerts on gaps in the log sequence

RPO achieved: < 1 minute in lab; < 5 seconds with continuous_log_interval=5.

Validated in E2E T09-F (base) + T09-I (continuous).

6.2 Point-in-Time Recovery (PITR)

podheitor-informix: mode=whole_server stop_at_time="2026-04-28 14:30:00"
                    point_in_time_tz=America/Sao_Paulo

Options:

• stop_at_lsn=<N> — stop log replay at specific LSN
• stop_at_time=<datetime> — stop at timestamp
• stop_before_txn=<txnid> — stop before a specific transaction (for targeted rollback)

The plugin replays the backup chain L0 → L1 → L2 → logical logs up to the chosen point, then runs onmode -m to bring the instance online. Validated in E2E T05.

6.3 Instant Recovery (FUSE Mount)

podheitor-informix: mode=instant instant_fuse_mount=/var/lib/podheitor-informix/instant

1. Restores chunk file headers and page 0 of each dbspace
2. FUSE filesystem presents virtual chunk files backed by lazy-streamed Bacula data
3. IDS boots via oninit -PHY against the FUSE-mounted chunks
4. Instance is queryable in < 60 seconds
5. Remaining blocks stream lazily in the background while IDS serves queries

Use case: immediate business continuity while full restore completes in background. Validated in E2E T17 using /dev/loop-control + /dev/fuse on VM 189.

6.4 Single-Table Extraction

podheitor-informix: mode=table_level table=app_db.dbo.orders
                    sandbox_docker_image=icr.io/informix/informix-developer-database:latest
                    single_item_export_dir=/tmp/restored

1. Spins up an ephemeral Podman/Docker Developer Edition container
2. Restores full backup chain into the sandbox
3. Replays to the requested time point
4. Runs dbexport/unload for the target table
5. Copies output to single_item_export_dir
6. Destroys the sandbox container

Result: Target table extracted in < 5 min from a 100 GB backup, zero production impact. Validated in E2E T07.

6.5 Chain Integrity Enforcement

By default (chain_health_check=yes), before every Differential or Incremental:

1. Plugin queries sysmaster:bar_action for the preceding level-0 LSN
2. Validates the manifest chain (SHA-256 of each sidecar)
3. Compares against Bacula catalog job history

If the chain is broken (missing job, LSN gap, SHA mismatch), the job aborts before writing any data, preventing a false-confidence backup. Operators are notified with a clear error explaining the gap.

6.6 Virtual Full (Synthetic Full)

virtual_full=yes triggers a server-side chain merge: the plugin reads L0+L1+L2 streams from Bacula, merges them into a new L0 stream that Bacula ingests as a fresh Full backup. Eliminates long restore chains (no need to apply Diff+Incr at restore time) while avoiding another full backup from the IDS host.

6.7 Bare-Metal Disaster Recovery

Every backup automatically captures a RestoreObject containing:

• onconfig (ONCONFIG file)
• sqlhosts (connectivity definitions)
• ixbar.<servernum> (ON-Bar emergency boot file)

When restoring to a fresh host with no prior IDS installation:

1. Bacula restores the RestoreObject first
2. Plugin reconstructs the IDS directory structure
3. Runs standard oninit initialization
4. Restores chunks and replays logs

Zero manual DBA intervention required. Validated in E2E T18.

---

7. Security & Compliance

7.1 Security Controls

Surface	Control	Default
Plugin configuration	passfile path verified to chmod 600; backend aborts if looser	Enforced
Connection strings in logs	Secret regex redaction in all PTCOMM logger output	redact_sysuser_passwords=yes
DBCRYPT keystore	Only encrypted blob (hash pointer) stored in __ph_dbcrypt.json; plaintext key NEVER logged or stored	dbcrypt_capture=yes
XBSA library load path	Verified against /opt/podheitor-informix/lib/; LD_PRELOAD rejected	Enforced
Working directory	/var/run/podheitor-informix mode 0700 owned by informix	Enforced
Sandbox container	Ephemeral, --read-only root + tmpfs; deleted after extraction	Enforced
FUSE mount	Read-only; mounted as informix:informix mode 0700	Enforced
ON-Bar exit codes	Non-zero triggers job fail	chain_health_check=yes
Correlation IDs	All PTCOMM messages structured-logged with Bacula JobId	Always

7.2 DBCRYPT Support

Informix’s DBCRYPT feature encrypts data at rest at the storage engine level. PodHeitor:

• Captures the keystore metadata (not the key) in __ph_dbcrypt.json
• Records the key label/alias for restore-side hook invocation
• On restore, invokes the DBCRYPT passphrase hook before oninit
• Result: encrypted-at-rest databases back up and restore with no manual key management

Validated in E2E T15.

7.3 ARD Ransomware Hook

When ransomware_hook=yes, the plugin runs onstat -g ARD before the backup and checks for:

• Entropy anomalies in recently written logical log blocks
• Sudden increase in UPDATE/DELETE rates (Active Ransomware Detection, IDS 14.10+)

If an anomaly is detected, the backup aborts with a clear warning, preventing the ransomware-encrypted state from overwriting a clean backup.

7.4 File Permission Hardening Checklist

# Verify correct permissions after install
ls -la /opt/bacula/plugins/podheitor-informix-fd.so    # -rwxr-xr-x root root
ls -la /opt/bacula/plugins/podheitor-informix-backend   # -rwxr-x--- root informix
ls -la /opt/podheitor-informix/lib/libpodheitor_xbsa.so # -rwxr-x--- root informix
ls -la /opt/bacula/etc/podheitor-informix.conf          # -rw------- informix informix
ls -la /var/run/podheitor-informix/                     # drwx------ informix informix

---

8. Observability

8.1 Prometheus Metrics

When metrics_endpoint=localhost:9100 is set, the plugin exposes a standard Prometheus /metrics endpoint.

Metric	Type	Description
podheitor_ifx_job_bytes_total	Counter	Total bytes transferred this job
podheitor_ifx_job_throughput_bytes_sec	Gauge	Current transfer rate
podheitor_ifx_dbspace_progress_pct	Gauge	Per-dbspace completion percentage
podheitor_ifx_logship_lag_seconds	Gauge	Log shipping lag (continuous_log mode)
podheitor_ifx_onbar_exit_code	Gauge	Last ON-Bar invocation exit code
podheitor_ifx_archecker_pass	Gauge	1=pass, 0=fail (last archecker run)
podheitor_ifx_queue_depth	Gauge	PTCOMM write-ahead queue depth
podheitor_ifx_active_stripes	Gauge	Current number of active PTCOMM streams
podheitor_ifx_xbsa_buffer_pool_used	Gauge	XBSA buffer pool utilization (bytes)

Example Grafana query for throughput:

rate(podheitor_ifx_job_bytes_total[1m])

Example Grafana query for log shipping lag alert:

podheitor_ifx_logship_lag_seconds > 300

8.2 OpenTelemetry OTLP Traces

When otel_endpoint=http://collector:4317 is set, the plugin emits structured spans to an OTLP-compatible collector (Jaeger, Tempo, Honeycomb, etc.).

Span hierarchy:

backup.job (job_id=NNNNN, instance=ifxprod)
  ├── engine.xbsa.init
  ├── dbspace.rootdbs.backup (level=0)
  │     ├── stripe.0.compress
  │     └── stripe.0.ptcomm_write
  ├── dbspace.app_data.backup (level=0)
  ├── manifest.write
  ├── archecker.verify
  └── topology.snapshot

---

9. Installation Guide

9.1 Prerequisites

Bacula File Daemon host (co-located with IDS):

• OS: OL9/RHEL 8/9/Rocky 8/9/AlmaLinux 8/9, Ubuntu 20.04/22.04/24.04, Debian 11/12, SLES 15
• PodHeitor Backup FD 15.0.3+
• HCL Informix 12.10.FC12+ / 14.10.FCx / 15.0.x installed
• informix user/group with standard IDS privileges
• For mode=snapshot: LVM or ZFS on dbspace devices, losetup+lvcreate available
• For mode=instant: /dev/fuse + /dev/loop-control accessible
• For mode=table_level: Podman or Docker installed

Bacula Director:

• Director 15.0.x with the FileSet and Job definitions (see §10)

9.2 Install on EL8/EL9 (RPM)

# Download the RPM from releases/
rpm -ivh podheitor-informix-plugin-0.1.0-1.el9.x86_64.rpm

# Verify installation
ls /opt/bacula/plugins/podheitor-informix-fd.so
ls /opt/bacula/plugins/podheitor-informix-backend
ls /opt/podheitor-informix/lib/libpodheitor_xbsa.so
ls /opt/bacula/etc/podheitor-informix.conf.sample

9.3 Install on Ubuntu 22.04 / Debian 12 (DEB)

# Download the DEB from releases/
dpkg -i podheitor-informix-plugin_0.1.0-1_amd64.deb

# Verify
dpkg -L podheitor-informix-plugin

9.4 Configure the Plugin

# Copy and edit the sample config
cp /opt/bacula/etc/podheitor-informix.conf.sample 
   /opt/bacula/etc/podheitor-informix.conf
chmod 600 /opt/bacula/etc/podheitor-informix.conf
chown informix:informix /opt/bacula/etc/podheitor-informix.conf

vi /opt/bacula/etc/podheitor-informix.conf

Minimum required settings:

[connection]
informixdir    = "/opt/ibm/informix"
informixserver = "ifxprod"
onconfig       = "onconfig.std"

[engine]
mode           = "auto"   # xbsa if onbar available, else ontape
compress       = "zstd"
compress_level = 3

9.5 Configure BAR Environment

# /etc/sysconfig/podheitor-informix  (read by onbar and backend)
cat > /etc/sysconfig/podheitor-informix << 'EOF'
INFORMIXDIR=/opt/ibm/informix
INFORMIXSERVER=ifxprod
ONCONFIG=onconfig.std
INFORMIXSQLHOSTS=/opt/ibm/informix/etc/sqlhosts
PATH=/opt/ibm/informix/bin:$PATH
BAR_BSALIB_PATH=/opt/podheitor-informix/lib/libpodheitor_xbsa.so
BAR_MAX_BACKUP=4
BAR_XFER_BUF_SIZE=1048576
BAR_NB_XPORT_COUNT=10
EOF
chmod 640 /etc/sysconfig/podheitor-informix
chown root:informix /etc/sysconfig/podheitor-informix

9.6 Preflight Check

/opt/bacula/plugins/podheitor-informix-backend --dry-run 
  /opt/bacula/etc/podheitor-informix.conf

Expected output:

[INFO] PodHeitor Informix v0.1.0 — preflight check
[INFO] IDS version: 15.0.1.0.3
[INFO] ON-Bar available: yes
[INFO] archecker available: yes
[INFO] XBSA library: /opt/podheitor-informix/lib/libpodheitor_xbsa.so [OK]
[INFO] BAR_BSALIB_PATH configured: yes
[INFO] Working dir: /var/run/podheitor-informix [created, mode 0700]
[INFO] Connectivity: ifxprod@localhost:9088 [OK]
[INFO] DRY-RUN COMPLETE — all checks passed

9.7 Restart bacula-fd

systemctl restart bacula-fd
journalctl -u bacula-fd -n 50 | grep -i podheitor
# Expected: "Loaded plugin: podheitor-informix v0.1.0"

---

10. Configuration Reference

10.1 Backup Options

#	Parameter	Default	Type	Description
Connection
1	informixdir	$INFORMIXDIR	path	IDS install directory
2	informixserver	$INFORMIXSERVER	string	Server name (sqlhosts entry)
3	onconfig	$ONCONFIG	file	onconfig file override
4	sqlhosts	$INFORMIXSQLHOSTS	file	sqlhosts file override
5	user	informix	string	Run-as OS user
6	passfile	—	path	Credential file (chmod 600)
7	kerberos_keytab	—	path	Kerberos keytab for Trusted Context
Engine
8	mode	auto	enum	`auto	xbsa	onbar_psm	ontape	logical_dbexport	snapshot	replicate_hdr	replicate_rss	replicate_sds	replicate_er`
Scope
9	dbspace	*	glob	Which dbspaces to back up
10	include_dbspace	—	glob (multi)	Explicit include list
11	exclude_dbspace	tempdb*	glob (multi)	Exclusion list
12	database	—	glob	Logical-mode DB filter
13	whole_system	yes	bool	Whole-server vs per-dbspace
14	include_blobspace	yes	bool	Traditional blobspaces
15	include_sbspace	yes	bool	Smart-blob sbspaces
16	include_external_dbspace	no	bool	External (read-only) dbspaces
17	include_temp_dbspace	no	bool	Temp dbspaces
Levels
18	level_full	0	int	Map Bacula Full → ON-Bar level 0
19	level_diff	1	int	Map Bacula Differential → level 1
20	level_incr	logical_log	enum	`2	logical_log`
21	force_full_after	7	int	Auto-promote to Full after N incrementals
22	chain_health_check	yes	bool	Verify chain before Diff/Incr
23	simulate_full	no	bool	Force level 0 regardless of Bacula request
24	copy_only	no	bool	Side-channel backup (doesn’t disturb chain)
Performance
25	bar_max_backup	# CPU cores	int	Sets BAR_MAX_BACKUP (parallel ON-Bar streams)
26	stripes	1	int	N-way striping per stream (1–32)
27	parallel_dbspaces	# CPU cores	int	Concurrent dbspace backups
28	bar_xfer_buf_size	1048576	bytes	Sets BAR_XFER_BUF_SIZE
29	bar_nb_xport_count	10	int	Sets BAR_NB_XPORT_COUNT
30	xbsa_buffer_count	8	int	XBSA shared buffer pool depth
31	xbsa_block_size	65536	bytes	XBSA block size
32	compress	zstd	enum	`none	zstd	lz4`
33	compress_level	3	int	ZSTD compression level (1–22)
34	lob_chunked_streaming	yes	bool	Content-defined chunking for sbspace LOBs
35	max_rate_kbps	0	int	Per-job bandwidth throttle (0=unlimited)
36	cpu_quota	0	pct	cgroup CPU quota for backend process
Log Shipping
37	continuous_log	no	bool	Background log shipping daemon
38	continuous_log_interval	60	seconds	Polling interval (min 5)
39	continuous_log_archive_only	yes	bool	Stop after BAR_RETRY failures
40	salvage_logs	no	bool	onbar -l -s salvage after crash
Integrity
41	oncheck_pre_backup	no	enum	`no	ce	cI	cD`
42	oncheck_post_backup	no	enum	`no	ce	cI	cD`
43	archecker_verify	yes	bool	Run archecker after backup
44	manifest_sha256	yes	bool	Per-stream SHA256 in manifest
45	verify_lsn_chain	yes	bool	Cross-check sysmaster:bar_action LSNs
Replication
46	topology_snapshot	yes	bool	Snapshot HDR/RSS/SDS/ER topology
47	er_subscribe_snapshot	yes	bool	Capture ER replicates + grids
48	hdr_role_check	any	enum	`primary	secondary	any`
49	prefer_secondary_for_backup	no	bool	Back up from HDR secondary
Security
50	dbcrypt_capture	yes	bool	Capture DBCRYPT keystore metadata
51	redact_sysuser_passwords	yes	bool	Filter sysmaster:sysuser from exports
52	audit_export	no	bool	Include audit trail
53	ransomware_hook	no	bool	Abort on ARD anomaly detection
Observability
54	metrics_endpoint	—	host:port	Prometheus /metrics endpoint
55	otel_endpoint	—	URI	OTLP traces collector
56	progress_report_interval	30	seconds	Job-log progress cadence
Operational
57	dry_run	no	bool	Preflight only (no write)
58	abort_on_chain_break	yes	bool	Refuse Diff/Incr if chain broken
59	working_dir	/var/run/podheitor-informix	path	FIFO + sentinel dir
60	config_file	/opt/bacula/etc/podheitor-informix.conf	path	External config TOML
61	cluster_id	—	string	Multi-cluster namespace separator
62	bar_ixbar_path	$INFORMIXDIR/etc/ixbar.<servernum>	path	ON-Bar emergency-boot file
63	restore_object	yes	bool	Store onconfig+sqlhosts+ixbar as RestoreObject
64	virtual_full	no	bool	Synthetic full from L0+L1+L2 chain

10.2 Restore Options

#	Parameter	Default	Description
1	mode	auto	`auto	whole_server	in_place	new_instance	single_dbspace	table_level	instant	ag_seed	logical_import`
2	informixdir	$INFORMIXDIR	Target IDS install directory
3	informixserver	(source name)	New IDS server name
4	onconfig	$ONCONFIG	Target onconfig file
5	dbspace_relocation_map	—	JSON file for logical→physical chunk path remapping
6	target_root_dbspace	—	Override rootdbs chunk path
7	target_dbspace	—	Single-dbspace mode target
8	dbspace	*	Which dbspaces to restore
9	database	*	Logical-mode DB filter
10	table (multi)	—	mode=table_level target tables
11	column_filter	—	Optional column-level filter (table_level)
12	recovery	yes	WITH RECOVERY (yes) vs NORECOVERY (for HDR seed)
13	stop_at_lsn	—	Stop log replay at LSN
14	stop_at_time	—	Stop log replay at timestamp
15	stop_before_txn	—	Stop before specific transaction
16	point_in_time_tz	UTC	Timezone for stop_at_time
17	physical_only	no	Restore physical only (skip logical log replay)
18	drop_existing	no	`no	yes	only_if_match`
19	tail_log_before	yes	Salvage current logical logs before destructive restore
20	verify_post	archecker	`no	oncheck	archecker	both`
21	dry_run	no	Preflight (validate manifest + paths, no write)
22	parallel_streams	(backup value)	Concurrent restore streams
23	stripes	(backup value)	Match backup stripes
24	parallel_dbspaces	# CPU cores	Concurrent dbspace restores
25	instant_fuse_mount	/var/lib/podheitor-informix/instant	FUSE mount for instant recovery
26	sandbox_docker_image	icr.io/informix/informix-developer-database:latest	Sandbox image for table_level
27	single_item_export_dir	/var/lib/podheitor-informix/extracted	Output for table_level
28	ag_target_role	—	`hdr_secondary	rss	sds`
29	ag_primary_host	—	Primary to attach to after seed
30	dbcrypt_keystore	—	Override DBCRYPT keystore location
31	dbcrypt_passphrase_file	—	Passphrase for sealed keystore
32	progress_report_interval	30	Progress cadence
33	regexwhere	—	Bacula standard regex relocation

10.3 FileSet Examples

Basic whole-server backup (XBSA, defaults):

FileSet {
  Name = "Informix-Full"
  Include {
    Options { Signature = SHA256 }
    Plugin = "podheitor-informix:"
  }
}

Specific dbspaces, striped, with verification:

FileSet {
  Name = "Informix-AppData"
  Include {
    Plugin = "podheitor-informix: include_dbspace=app_data include_dbspace=app_index
              exclude_dbspace=tempdb* archecker_verify=yes compress=zstd
              stripes=4 parallel_dbspaces=2"
  }
}

Continuous log shipping (sub-minute RPO):

FileSet {
  Name = "Informix-LogShip"
  Include {
    Plugin = "podheitor-informix: continuous_log=yes continuous_log_interval=30"
  }
}

Cross-version migration:

FileSet {
  Name = "Informix-Export"
  Include {
    Plugin = "podheitor-informix: mode=logical_dbexport database=app_db"
  }
}

Snapshot mode (LVM):

FileSet {
  Name = "Informix-Snapshot"
  Include {
    Plugin = "podheitor-informix: mode=snapshot snapshot_vg=ifxvg snapshot_lv=ifxdata snapshot_size=1G"
  }
}

VLDB high-performance backup (8 stripes, 4 parallel dbspaces):

FileSet {
  Name = "Informix-VLDB"
  Include {
    Plugin = "podheitor-informix: stripes=8 parallel_dbspaces=4 bar_max_backup=8
              bar_xfer_buf_size=4194304 compress=zstd compress_level=3"
  }
}

Backup from HDR secondary (offload primary):

FileSet {
  Name = "Informix-HDR-Secondary"
  Include {
    Plugin = "podheitor-informix: prefer_secondary_for_backup=yes hdr_role_check=secondary"
  }
}

10.4 Job Definitions

Job {
  Name     = "informix-prod-full"
  JobDefs  = "DefaultJob"
  Type     = Backup
  Level    = Full
  FileSet  = "Informix-Full"
  Schedule = "WeeklyFull"
  Client   = "ifx-prod-fd"
  Storage  = "DiskStorage"
  Pool     = "Informix-Full"
}

Job {
  Name     = "informix-prod-diff"
  JobDefs  = "DefaultJob"
  Type     = Backup
  Level    = Differential
  FileSet  = "Informix-Full"
  Schedule = "DailyDiff"
  Client   = "ifx-prod-fd"
  Storage  = "DiskStorage"
  Pool     = "Informix-Diff"
}

Job {
  Name        = "informix-prod-logship"
  JobDefs     = "DefaultJob"
  Type        = Backup
  Level       = Incremental
  FileSet     = "Informix-LogShip"
  Schedule    = "Every5Min"
  Client      = "ifx-prod-fd"
  Storage     = "DiskStorage"
  Pool        = "Informix-Logs"
  Max Run Time = 10 minutes
}

---

11. Restore Procedures & Use Cases

11.1 Whole-Server Restore (In-Place)

bconsole> restore client=ifx-prod-fd
[browse and mark]
bconsole> mod
Plugin Options: podheitor-informix: mode=in_place drop_existing=only_if_match
                tail_log_before=yes verify_post=both
bconsole> yes

11.2 Restore to New Instance (Disaster Recovery)

bconsole> restore client=ifx-prod-fd where=/
Plugin Options: podheitor-informix: mode=new_instance informixserver=ifxprod-dr
                dbspace_relocation_map=/etc/podheitor-informix/dr-relocation.json
                verify_post=archecker

11.3 Point-in-Time Recovery

Plugin Options: podheitor-informix: mode=whole_server
                stop_at_time="2026-04-28 14:30:00"
                point_in_time_tz=America/Sao_Paulo
                verify_post=both

11.4 Instant Recovery

Plugin Options: podheitor-informix: mode=instant
                instant_fuse_mount=/var/lib/podheitor-informix/instant

Expected timeline:

• 0–15s: restores chunk headers + page 0
• 15–60s: IDS boots against FUSE mount
• 60s+: instance online, lazy-streaming continues

11.5 Single-Table Extraction

Plugin Options: podheitor-informix: mode=table_level
                table=app_db.dbo.orders
                stop_at_time="2026-04-28 12:00:00"
                sandbox_docker_image=icr.io/informix/informix-developer-database:latest
                single_item_export_dir=/tmp/orders-restored

11.6 HDR Secondary Bootstrap

Plugin Options: podheitor-informix: mode=ag_seed
                ag_target_role=hdr_secondary
                ag_primary_host=ifx-prod:9088
                recovery=no

11.7 Cross-Version Migration (12.10 → 15.0)

Step 1 — backup source (12.10):

Plugin = "podheitor-informix: mode=logical_dbexport database=app_db"

Step 2 — restore to 15.0 target:

Plugin Options: podheitor-informix: mode=logical_import informixdir=/opt/ibm/informix
                informixserver=ifxprod-new database=app_db

---

12. Performance Reports

12.1 Throughput Benchmarks (Lab — OL9, IDS 15.0.1, 1 Gbps network)

DB Size	Mode	stripes	parallel_dbspaces	Throughput	Wall Time
10 GB	xbsa (zstd-3)	2	1	250 MB/s	42 s
10 GB	ontape (zstd-3)	1	n/a	85 MB/s	2 min
50 GB	xbsa (zstd-3)	4	2	420 MB/s	2 min
100 GB	xbsa (zstd-3)	4	4	440 MB/s	4 min
100 GB	snapshot (zstd-3)	4	4	520 MB/s	3.5 min

12.2 Compression Ratio Comparison

Data Type	zstd-3	zstd-9	lz4	none
OLTP (mixed)	2.8×	3.5×	2.2×	1×
Text/documents	6.2×	8.0×	4.5×	1×
IDS system dbspaces	3.1×	4.0×	2.4×	1×
Binary BLOBs	1.05×	1.05×	1.02×	1×

Recommendation: compress=zstd compress_level=3 for most environments. Use lz4 for CPU-constrained hosts; zstd compress_level=9 for backup storage cost reduction.

12.3 RPO / RTO Targets

Configuration	RPO	RTO
Daily Full only	24 h	1–4 h
Weekly Full + daily Diff	24 h	30 min–2 h
Weekly Full + daily Diff + 5-min log ship	< 5 min	1–3 h
Continuous log shipping (default)	< 1 min	1–3 h
Continuous log shipping (interval=5)	< 5 s	1–3 h
HDR secondary + nightly backup	seconds	< 5 min (failover)
Instant recovery (FUSE)	as backup	< 60 s
Snapshot + log shipping	< 30 s	30 min–2 h

12.4 Memory Footprint

Default configuration: xbsa_buffer_count=8 × xbsa_block_size=64KiB × stripes=1 × parallel_dbspaces=# cores

stripes	parallel_dbspaces	Memory ceiling
1	1	0.5 MB
4	4	8 MB
8	8	32 MB
32	16	256 MB

FD host disk footprint: < 50 MB (FIFOs + sentinels + manifest only). No staging files ever written.

---

13. Compatibility Matrix

13.1 IDS Version × Feature Matrix

Feature	IDS 12.10.FC12	IDS 14.10	IDS 15.0
mode=xbsa	✅	✅	✅
mode=ontape	✅	✅	✅
mode=onbar_psm	✅	✅	✅
mode=logical_dbexport	✅	✅	✅
mode=snapshot	✅	✅	✅
Continuous log shipping	✅	✅	✅
PITR (stop_at_lsn)	✅	✅	✅
HDR bootstrap	✅	✅	✅
RSS bootstrap	✅	✅	✅
ARD ransomware hook	❌ (< 14.10)	✅	✅
DBCRYPT capture	✅	✅	✅
Virtual Full	✅	✅	✅
Instant recovery (FUSE)	✅	✅	✅

13.2 OS × Package Availability

OS	RPM (el9)	DEB	Notes
Oracle Linux 9	✅	—	Primary GA target
RHEL 9 / Rocky 9 / AlmaLinux 9	✅	—	GA
RHEL 8 / Rocky 8	✅	—	GA (el9 RPM compatible)
Ubuntu 22.04 LTS	—	✅	GA
Ubuntu 20.04 LTS	—	✅	GA
Debian 12	—	✅	GA
Debian 11	—	✅	GA
SLES 15	✅	—	Via el9 RPM
AIX 7.2/7.3	—	—	Roadmap v0.2.0
Windows Server 2019/2022	—	—	Roadmap v0.3.0
Linux ARM64	—	—	Roadmap v0.3.0

13.3 Bacula Compatibility

PodHeitor Backup	Support
15.0.3	✅ Primary GA target
15.0.x	✅
14.x	✅ (PTCOMM compatible)
< 14	⚠️ Not tested

---

14. Sizing Guide

14.1 Minimum Requirements

Component	Minimum	Notes
FD host RAM	1 GB	XBSA buffer pool at defaults ≈ 0.5 MB
FD host CPU	2 cores	Compression + 1 parallel stream
FD host disk	50 MB	Working dir (FIFOs + manifest only)
FD host OS disk	—	No staging; never writes data to local disk
Network FD→SD	100 Mbps	Minimum functional

14.2 Recommended Production Sizing

IDS Environment	FD RAM	FD CPU	Network	stripes	parallel_dbspaces
Small (< 100 GB, single dbspace)	4 GB	4 cores	1 Gbps	2	1
Medium (100 GB–1 TB, 4–8 dbspaces)	8 GB	8 cores	10 Gbps	4	4
Large (1–10 TB, 10+ dbspaces)	16 GB	16 cores	25 Gbps	8	8
VLDB (> 10 TB, many dbspaces)	32 GB	32 cores	25+ Gbps	16–32	16+

14.3 Bacula Storage Daemon Sizing

Daily change rate	Storage pool	SD disk I/O	Notes
< 50 GB/day	4 TB disk	SAS/SATA	Standard pools
50–500 GB/day	20 TB disk	SSD recommended	Enable zstd compression on SD too
> 500 GB/day	NAS/SAN or tape	High throughput	Consider Bacula Dedup SD plugin

---

15. Bacularis Integration

Bacularis provides a web UI on top of PodHeitor Backup. The PodHeitor Informix plugin ships a bacularis/plugin-descriptor.json that:

• Adds “Informix” as a client resource type in the Bacularis backup wizard
• Renders all 64 backup parameters as form fields with validation
• Shows the restore browser with IDS-aware namespace (/@informix/) tree view
• Displays E2E validation results inline

To enable the descriptor:

cp /opt/bacula/plugins/bacularis/plugin-descriptor.json 
   /var/www/bacularis/protected/Common/Modules/PluginConfig/podheitor-informix.json
systemctl restart bacularis

---

16. Migration from TSM/ISP and ON-Bar PSM

16.1 Migrating from IBM Spectrum Protect (TSM/ISP)

See MIGRATION_FROM_TSM.md for the full procedure. Summary:

1. Install PodHeitor Backup + PodHeitor on the IDS host
2. Run parallel backups (TSM + PodHeitor) for one full backup cycle to validate
3. Confirm chain integrity via archecker + manifest
4. Set BAR_BSALIB_PATH to point to PodHeitor library (replaces TSM library)
5. Decommission TSM client

Key difference: TSM’s BAR_BSALIB_PATH points to its own library. PodHeitor’s library is a drop-in replacement — ON-Bar doesn’t know the difference.

16.2 Migrating from ON-Bar PSM Scripts

See MIGRATION_FROM_PSM.md. Summary:

1. Identify your current BAR_MAX_BACKUP, BAR_XFER_BUF_SIZE, backup windows
2. Create equivalent PodHeitor FileSet with mode=onbar_psm as a transition step
3. Validate one full backup cycle
4. Switch to mode=xbsa (default) for direct streaming
5. Remove PSM scripts and staging disk allocation

Storage savings from removing staging: PSM writes to disk (.bak files) before Bacula picks them up. PodHeitor streams directly, eliminating 1× backup set of disk space.

---

17. Troubleshooting

17.1 Plugin Not Loaded

bacula-fd: error: plugin "podheitor-informix" not found

Cause: .so not in Bacula plugin directory. Fix: Verify /opt/bacula/plugins/podheitor-informix-fd.so exists; check bacula-fd.conf has Plugin Directory = /opt/bacula/plugins.

17.2 XBSA Library Load Failure

[ERROR] BAR_BSALIB_PATH not set or library not found

Cause: BAR_BSALIB_PATH not set in /etc/sysconfig/podheitor-informix. Fix:

echo "BAR_BSALIB_PATH=/opt/podheitor-informix/lib/libpodheitor_xbsa.so" 
  >> /etc/sysconfig/podheitor-informix
systemctl restart bacula-fd

17.3 ON-Bar Exits Non-Zero

[ERROR] onbar exit code 45 — see /tmp/onbar_<jobid>.log

Cause: ON-Bar error. Check the log file. Fix: Typically: wrong INFORMIXSERVER, IDS not running, or sysmaster.bar_action locked.

cat /tmp/onbar_<jobid>.log
onstat -g ses   # check active IDS sessions

17.4 Chain Health Check Failure

[ABORT] Chain broken: L0 JobId=1234 not found in catalog (or SHA mismatch)

Cause: Differential/Incremental job aborted because the preceding Full or Diff is missing. Fix: Run a new Full backup. If the Full exists and SHA is wrong, the backup volume may be corrupt — investigate SD device.

17.5 Continuous Log Shipping Gap

[WARN] Log gap detected: expected log unique_id=150, got 152

Cause: Log file 151 was never closed (IDS bug) or was already overwritten. Fix:

onbar -l -s   # salvage logs from disk
# OR set salvage_logs=yes in config

17.6 Diagnostic Commands

# Check IDS status
onstat -

# Check ON-Bar backup history
onstat -g bac

# Check logical log status
onstat -l

# Run archecker manually on last backup
archecker -b -v -f /@informix/<UUID>/__ph_manifest.json

# Check plugin runtime
journalctl -u bacula-fd -n 200 | grep -i podheitor

# Dry-run preflight
/opt/bacula/plugins/podheitor-informix-backend --dry-run 
  /opt/bacula/etc/podheitor-informix.conf

---

18. Roadmap

Version	Target	Key Features
v0.1.0 GA	2026-04-29	Full Linux x86_64 support, all 22 E2E scenarios PASS
v0.2.0	2026 Q3	AIX 7.x support (ppc64), NetApp ONTAP snapshot provider, OneDB 2.x compatibility
v0.3.0	2026 Q4	Windows Server FD, Linux ARM64, ER multi-mesh restore
v0.4.0	2027 Q1	Bacularis UI plugin form descriptor, Kubernetes sidecar FD
v0.5.0	2027 Q2	Cross-version logical apply replication (12.10 → 14.10 live streaming)
v1.0.0	2027 Q3	Full platform support (Linux + AIX + Windows + ARM64) + Bacularis UI

---

19. Validation Evidence (E2E Results)

19.1 Lab Environment

Component	Details
Bacula Director	192.168.15.105, Community 15.0.3, OL9
IDS Primary (VM 186)	192.168.15.130, ifxprod, IDS 15.0.1.0.3 Developer Edition in Docker
IDS HDR Secondary (VM 187)	192.168.15.131, ifxhdr, IDS 15.0.1.0.3
IDS RSS Secondary (VM 188)	192.168.15.132, ifxrss, IDS 15.0.1.0.3
IDS Snapshot Host (VM 189)	192.168.15.133, ifxsnap, OL9 bare-metal + loop-backed LVM VG (ifxvg) + FUSE3

19.2 E2E Results — 2026-04-29 (JobIds 6358–6379)

ID	Scenario	JobId	Level	Result
T01	Whole-server Full backup	6358	Full	✅ PASS
T02-F	Full+Diff+Incr chain — Full	6359	Full	✅ PASS
T02-D	Full+Diff+Incr chain — Differential	6360	Differential	✅ PASS
T02-I	Full+Diff+Incr chain — Incremental	6361	Incremental	✅ PASS
T03	Restore in-place (restore_object=yes)	6362	Full	✅ PASS
T04	Restore to new instance	6363	Full	✅ PASS
T05-F	PITR chain — Full baseline	6364	Full	✅ PASS
T05-I	PITR chain — Incremental (log-ship)	6365	Incremental	✅ PASS
T06	Single-dbspace (rootdbs only)	6366	Full	✅ PASS
T07	Table-level backup & extract	6367	Full	✅ PASS
T08	4-stripe parallel backup	6368	Full	✅ PASS
T09-F	Continuous log shipping — base	6369	Full	✅ PASS
T09-I	Continuous log shipping — Incremental	6370	Incremental	✅ PASS
T10	HDR secondary bootstrap	6371	Full	✅ PASS
T11	RSS secondary bootstrap	6372	Full	✅ PASS
T12	Logical export (mode=logical_dbexport)	6373	Full	✅ PASS
T13	LVM snapshot (snapshot_vg=ifxvg)	6374	Full	✅ PASS
T14	HDR seeding via Bacula transport	6375	Full	✅ PASS
T15	DBCRYPT metadata sidecar	6376	Full	✅ PASS
T16	ontape explicit fallback	6377	Full	✅ PASS
T17	Instant recovery (loopdev+FUSE)	6378	Full	✅ PASS
T18	Bare-metal RestoreObject DR	6379	Full	✅ PASS

PASS=22 / FAIL=0 / SKIP=0 / TOTAL=22 — 100% clean sweep

---

Contact & Licensing

Licensing

PodHeitor Informix is proprietary software distributed by subscription. For commercial terms, technical demo, or a free 30-minute assessment, reach the team via the channels below.

Ready to evaluate?

• 💬 WhatsApp: +1 (786) 726-1749
• ✉️ Email: heitor@opentechs.lat
• 🩺 Free assessment — 30 min with Heitor Faria

---

PodHeitor Informix Backup, Replication and Conversion Plugin for Bacula — v0.1.0 GA © 2026 Heitor Faria. All rights reserved.

Disponível em: Português (Portuguese (Brazil))EnglishEspañol (Spanish)