Intermittent Failure Capture

Capture evidence during irregular failure episodes so elusive problems can be diagnosed after the episode disappears.

Solution archetype #

559

Problem family

Observability, Measurement & Feedback Gaps

Problem subfamily

Temporal Sampling, Decay & Transition Resolution

The Diagnostic Story

Symptom: The failure appears, the system is checked, the failure has disappeared. The investigation finds nothing because the failure state was erased by recovery, cleanup, or the simple passage of time. Anecdotal episode reports accumulate but diagnosis never advances because no two descriptions quite match and there is no evidence from the actual failure moment. The same unknown root cause recurs at irregular intervals with no path to resolution.

Pivot: Instrument the system, process, or human reporting workflow so that when an intermittent episode occurs, the relevant state, context, sequence, and traces are captured inside a protected window before evidence disappears, then routed into a diagnostic follow-up path that persists after normal operation resumes.

Resolution: Episodes become diagnosable because the evidence survives the failure, and root cause ambiguity decreases as captured context links state to sequence. Corrective action becomes possible and better calibrated because it is based on observed failure conditions rather than reconstructed guesses. Diagnosis is less disruptive because the capture mechanism does the work during the episode rather than requiring a recreation attempt later.

Reach for this when you hear…

[automotive engineering] “The check engine light comes on during the commute and goes off before I reach the dealer — we need a freeze-frame capture in the ECU so the code is still there when we plug in the scanner.”

[site reliability engineering] “The error rate spike lasts four minutes and our metrics retention is five-minute rollups — we're throwing away the exact window we need to diagnose it.”

[clinical neurology] “The patient's seizures happen twice a month for thirty seconds — we need a wearable EEG that records continuously and flags the episode, because they'll never be seizing in my office.”

When This Archetype Applies

Partial catalog groundingSome structural conditions are represented by existing abstractions, but no sufficient condition set is fully represented.

Diagnostic problem

A failure, symptom, anomaly, or breakdown appears irregularly and cannot be reliably reproduced or inspected during ordinary investigation, so the system repeatedly returns to normal before enough evidence is captured to explain what happened.

What this problem means

The structural problem is an evidence timing mismatch. Diagnosis needs information from the failure state, but the failure state is intermittent and self-erasing. Logs rotate, systems reset, symptoms fade, witnesses forget, cleanup routines remove evidence, and people reinterpret episodes after the fact.

This creates recurring root-cause ambiguity. Each episode is treated as isolated or anecdotal because the common sequence across episodes is unavailable. The system may accumulate reports without accumulating diagnosis.

Show the applicability expression

Applicability expression4 distinct conditions

Sparse episodic failureandInspection misses episodesandEpisode context requiredandEphemeral diagnostic evidence

Algebraic1∧2∧3∧4

groundedpartly groundedopen

4 conditions, all required.

4Required in every casenumbered 1–4

These hold no matter which pattern applies.

1

Sparse episodic failure · grounded

Failure appears in sparse, irregular, high-deviation episodes separated by ordinary behavior. more

and — also required

2

Inspection misses episodes · grounded · any one of 2

Ordinary inspection is unlikely to coincide with the failure episode. more

and — also required

3

Episode context required · open

The episode's local sequence, state, and environment are necessary for diagnosis. more

and — also required

4

Ephemeral diagnostic evidence · grounded · any one of 2

Diagnostic evidence from the episode disappears or degrades rapidly. more

Other requirements and context (1)

Why these sit outside the expression

Solution feasibility — it describes whether the intervention can work, not whether the diagnostic problem exists.

• Solution feasibilityCapture is possible without unacceptable harm. more

Coverage

3 of 4 conditions grounded · 1 open.

Read the methodology · Download the trigger-logic data

Mechanisms / Implementations

• Automatic Diagnostic Capture: Automatically saves diagnostic details when a trigger condition is met, reducing dependence on the investigator being present.
• Black Box Log: Maintains tamper-resistant or failure-resistant traces for situations where ordinary records may be lost during the event itself.
• Flight Recorder: Continuously or selectively retains recent operational state so evidence is available when a rare episode occurs.
• Incident Snapshot: Packages state, context, logs, timestamps, and human observations into a preserved record for later analysis.
• Post-Episode Evidence Review: Reviews captured evidence soon after the episode while context is still recoverable and corrective action is still timely.
• Rare Event Monitor: Watches for low-frequency events and preserves evidence when they occur rather than relying on continuous human attention.
• Symptom Diary: Lets a person record timing, context, preceding conditions, symptoms, and recovery details for episodic problems that are not directly instrumented.
• Trigger-Based Debug Trace: Turns on richer logging or tracing when a symptom, threshold, or anomaly indicates that an intermittent episode may be occurring.

Related Abstractions

Abstractions this archetype builds on — directly (a source ingredient) or as a related pattern. Links follow the typed catalog namespace.

Built directly on (2)

• Intermittency: Irregular bursts.
• Observability: Infer internal state externally.

Also references 11 related abstractions

• Accountability: Responsibility for actions.
• Causality: Cause-effect relationships.
• Fault Tolerance: Continue operating under failure.
• Feedback: Outputs influence inputs.
• Irreversibility: Cannot revert state.
• Resilience: Absorb shocks and adapt.
• Sampling (Representativeness): Representative subset selection.
• Signaling: Revealing hidden information.
• State and State Transition: Captures system condition and evolution.
• Threshold: Safe vs harmful levels.

▸ Show 1 more

Variants

Narrower or domain-specific specializations that share this archetype's core structure. Recognized variants are established; candidate variants are provisional.

Trigger-Based Diagnostic Capture · recognized

Episodic Symptom Capture · recognized

Flight-Recorder Failure Capture · recognized

Rare Event Evidence Capture · recognized

Editorial Notes

Problem Classification

Classification: Observability, Measurement & Feedback Gaps → Temporal Sampling, Decay & Transition Resolution

Problem kernel: intermittent failure vanishes before investigation

Rationale: The system returns to normal between episodes, so ordinary observation never captures the short-lived state needed for diagnosis.

Independent corroboration: The earliest necessary condition in the frozen evidence is: A failure, symptom, anomaly, or breakdown appears irregularly and cannot be reliably reproduced or inspected during ordinary investigation, so the system repeatedly returns to normal before enough evidence is captured to explain what happened. That is a temporal sampling decay and transition resolution problem because Observation cadence or anchoring does not match state dynamics, so brief events, decaying effects, accumulated drift, and approaching transitions are missed.

Review outcome: Independent reviewer agreement; high confidence.