Reference

Troubleshooting

Diagnose common target-connection, execution, scoring, and result problems without losing the original evaluation evidence.

Start with the earliest failing stage. A target connection problem, case-generation problem, and scoring problem can produce similar symptoms in the final view, but they require different fixes.

Use the diagnostic sequence

  1. Identify the failing stage

    Determine whether the run failed during target validation, case preparation, target execution, evaluator scoring, or aggregation.

  2. Read the narrowest available error

    Prefer a case-level or stage-level message over a generic evaluation status. Record the time and any request or case identifier shown.

  3. Check what changed

    Compare target version, credentials, endpoint, deployment status, rate limits, dataset, and evaluation settings with the last successful run.

  4. Test the smallest safe unit

    Validate target connectivity or retry a limited test before launching the complete evaluation again.

  5. Create a traceable retry

    Keep the method and evidence unchanged when testing an infrastructure fix. If the target or method changed, treat the retry as a new evaluated scope.

Target connection problems

SymptomLikely causeCheck
Authentication or unauthorized errorMissing, expired, or incorrectly scoped credentialCredential validity, header format, secret scope, and target provider
Target not foundIncorrect model, deployment, route, or base URLExact deployment name and final request URL
Connection timeoutTarget unavailable, private network boundary, or timeout too lowService health, network access, response latency, and timeout setting
Empty or unreadable responseResponse shape differs from configured mappingRaw response structure and expected response field
Works in a manual client but not AIDXHeaders, proxy, TLS, origin, or network path differsCompare the complete request path without exposing credentials

Execution problems

SymptomInterpretationRecommended action
Run remains queuedNo case has begun target executionConfirm the run was started, then check platform status and workspace concurrency.
Progress stops after some casesA repeated target or rate-limit failure may be blocking executionInspect the latest failed cases and target service limits before retrying.
Many timeout casesTarget latency exceeds the execution window or the service is overloadedStabilize the target, review rate and timeout settings, then create a controlled retry.
Rate-limit errorsThe target rejects the evaluation request rateReduce concurrency or request rate when supported, or increase the target quota.
Responses differ unexpectedlyThe target is stochastic or changed during the runConfirm model/version and sampling settings; do not combine evidence from changed targets.

Scoring and result problems

A run can complete target execution while evaluation scoring is still incomplete. Separate missing responses from missing evaluator outcomes.

SymptomCheck firstWhy it matters
No headline scoreWhether all required cases and evaluator tasks completedAggregation may wait for every required outcome.
Unexpectedly low scoreDirectionality, included cases, and dominant weak categoriesA valid aggregate can still be driven by one concentrated risk area.
Score changed after retryTarget version, sampling, dataset, method, and exclusionsThe two runs are comparable only when their scope is equivalent.
Cases marked unscoredResponse readability and evaluator availabilityUnscored cases should not silently be treated as safe outcomes.

When to contact AIDX

Contact AIDX when the target connection succeeds but platform processing repeatedly fails, a completed run cannot be scored, or the evidence shown is internally inconsistent.

Include the following non-secret information:

IncludeDo not include
Workspace or organization namePasswords, API keys, access tokens, or full secret values
Evaluation name or identifierSensitive customer prompts unless an approved support channel is used
Target type and declared versionUnredacted personal or regulated data
Failing stage, status, timestamp, and error textCredentials embedded in request examples or screenshots
Whether a retry produced the same outcomeEdited result data presented as original evidence