Turning logs into long-term intelligence.
Oninit® DrWatson — Features
Ingest
- Reads any extracted ifxcollect directory — tarball extraction is the operator's responsibility.
- Detects the collection profile (-c <category> -s <subcategory>) from manifest.log.<TS> when present, otherwise infers it from which artefacts exist.
- Per-iteration awareness for ifxcollect -r N -d S multi-run collections (loader emits collection_iteration/4 facts).
Collectors (subset)
- onconfig — full parameter set plus BUFFERPOOL / VPCLASS structured rows.
- online.log — assertion failures, restarts, checkpoint timing, archive lifecycle.
- oninit -V / -version — instance version, server name, edition.
- onstat -d / -D — dbspaces and chunks parsed by hex flags (catches mirror, offline, downstate, archive flags).
- onstat -p — full profile counter set with per-iteration deltas.
- onstat -l — physical and logical log status, free/full LRU thresholds.
- onstat -g arc — per-dbspace archive recency against backup hygiene rules.
- onstat -g ckp — checkpoint history with duration, wait, reason.
- onstat -g iof — per-chunk I/O latency (read/write avg, max).
- onstat -g glo — MT global / VP class CPU breakdown.
- onstat -g ses — session list with memory and state.
- onstat -g osi — OS, shared memory and semaphore info.
- onstat -g env — server start-up environment.
Rules engine
- Rules authored as plain SWI-Prolog .pl files for review and hand-editing.
- Janus bridge gives clean bidirectional Python ↔ Prolog calls.
- assertz / retract rebuild the fact base per collection — no carry-over between runs.
- Meta-interpreter support makes "explain why this finding fired" tractable: each finding can return the chain of rule clauses and facts that justified it.
- Rules guard on profile presence so a narrow collection does not generate false negatives from rules whose data sources are absent.
- Severity tiers: critical, high, medium, low, info; emitted as finding/5 terms with rule id, title, severity, evidence and remediation.
Diagnostic coverage
| Family | Rule IDs | What it covers |
|---|---|---|
| Configuration audit | CFG-001..010 | onconfig parameters checked against best practice and version-specific guidance; BUFFERPOOL and VPCLASS rows analysed structurally per page-size class. |
| Health and errors | HLTH-001..011 AF-001..AF-008 |
backup hygiene leveraging the 0x04000000 archive-pending flag; assertion-failure pattern recognition; server restarts; online.log timeline reconstruction. AF rules detect crash signatures, NULL-pointer dereferences, recurring crashes, HDR/RSS log-replay corruption, and unknown-build versions. |
| Capacity and space | CAP-001..008 | dbspace fullness (correctly skipping plogspace and temp); logical-log fullness; backup pile-ups. |
| Performance and tuning | PERF-001..014 | buffer hit ratio; profile counters; per-chunk I/O latency from onstat -g iof; CPU VP saturation from onstat -g glo; lock and latch contention; session memory. |
Reporting
- Branded PDF report with embedded logo and multi-line footer block.
- Methodology appendix (opt-in) explaining how each rule derives its conclusion.
- Plain CLI report for headless / pipeline use — same renderer as the PDF.
- Recommendations section integrates DrWatson output: current value, suggested value, triggering rule(s), how to apply (dynamic via onmode -wf, restart required, or operator action).
DrWatson
| Family | Rule IDs | What it tunes |
|---|---|---|
| Backup | TUNE-BAK-001..004 | archive cadence, backup window, log retention. |
| Locks | TUNE-LOK-001..002 | LOCKS / DEF_TABLE_LOCKMODE against observed contention. |
| Bufferpool | TUNE-BUF-NNN | per page-size pool: size and LRU thresholds derived from observed hit ratio and turnover. |
The DrWatson suggests; an operator applies. Nothing is changed automatically.
Turning logs into long-term intelligence.