Troubleshoot Provider Connections

Failure categories

Provider failures usually come from one of four places: invalid credentials, provider account or billing limits, unsupported model access, or hosting-level connectivity issues. Do not rotate keys repeatedly until diagnostics show which category is likely.

Connection evidence

Record the provider name, timestamp, exact error, PHP version, extension status, outbound HTTPS result, and whether the provider account works outside WordPress. The provider troubleshooting tutorial shows the practical evidence-gathering workflow.

Escalation

Escalate to the host when extensions, DNS, SSL, or outbound connections fail. Escalate to the provider when billing, account limits, key restrictions, or model availability are the likely issue. Use diagnostics and support to keep the report safe.

Quick reference

• Use this page when collecting diagnostics, reproducing errors, planning updates, routing support, or documenting incident recovery.
• Do not share raw logs, provider keys, server credentials, purchase codes, payment data, or unrelated customer records as support evidence.
• Key decision: whether the evidence is fresh, redacted, reproducible, routed to the right owner, and safe to share.

Scope limits

• Do not use this page to share raw logs, credentials, purchase codes, or unredacted customer records while trying to accelerate support.
• This page does not replace hosting, provider, CodeCanyon, or internal incident ownership.
• Use it to collect safe evidence and route the issue without exposing secrets or unrelated private records.

Owner and cadence

• Primary owner: support lead or site administrator responsible for triage and evidence handling.
• Review cadence: when an issue is reported, before support contact, and after recovery to improve the runbook.
• Escalate when evidence is incomplete, private data may be exposed, production impact continues, or routing between host, provider, and plugin support is unclear.

Access and data boundary

• Support evidence should preserve exact error codes, timestamps, URLs, user agents, diagnostics, and reproduction paths while removing secrets and unrelated private data.
• Use support-safe diagnostics, exact timestamps, reproduction steps, version details, and affected-screen context instead of raw logs or broad database exports.
• Share provider keys, server credentials, purchase codes, payment data, customer records, and private logs only through approved secure channels when explicitly required.

Production checklist

• Check key validity, provider billing, model access, PHP extensions, outbound HTTPS, DNS, SSL, firewall, and host restrictions.
• Test whether the provider account works outside WordPress before rotating keys repeatedly.
• Capture exact timestamp, affected user, affected screen, SophMate version, WordPress version, PHP version, and reproduction steps.
• Redact provider keys, credentials, payment data, private customer details, and raw logs before support handoff.

Acceptance checks

• The failure is narrowed to credentials, provider account, model access, or hosting connectivity.
• The next escalation includes exact error text, timestamp, provider, and diagnostics evidence.
• The support report lets another operator reproduce or triage the issue without receiving secrets.
• The team knows whether to escalate to hosting, provider support, CodeCanyon support, or internal operations.

Failure modes to test

• Test repeated errors, stale diagnostics, failed migration or reindexing, missing purchase context, untested rollback, wrong support route, and incomplete incident prevention notes.
• Test stale diagnostics, repeated error codes, missing reproduction steps, unredacted evidence, wrong escalation route, failed update, and incomplete rollback proof.
• Confirm support handoff preserves timestamps and codes while removing secrets and unrelated private data.

Evidence to capture

• Record exact timestamp, screen, URL, user role, SophMate version, environment details, error code, reproduction path, and diagnostics status.
• Capture what was redacted from logs, screenshots, support bundles, and customer-related evidence.

Decision record

• Decision field to include: exact error or release context, redaction status, rollback proof, support route, prevention owner, and restart criteria.
• Record the support decision, severity, affected screen, timestamp, environment details, reproduction path, owner, routing path, and redaction status.
• Include whether the issue belongs to hosting, provider support, CodeCanyon/plugin support, internal operations, or a post-incident prevention task.

Stop or rollback path

Stop retries, updates, or support handoffs when evidence is incomplete, secrets are exposed, ownership is unclear, or rollback is untested. Resume after diagnostics, redaction, support route, and prevention notes are complete.

Monitoring window

• Monitor repeat errors, unresolved diagnostics, support response state, redaction quality, and owner handoff until closure.
• Review prevention notes after the issue is resolved so the next request is easier to triage.

Expansion criteria

• Standardize the support path only after reproduction, diagnostics, redaction, routing, and prevention notes are complete.
• The next operator can use the same report format without exposing secrets or unrelated customer data.

Common mistakes

• Retrying or changing settings repeatedly before preserving the exact error report, timestamp, and affected screen.
• Opening support requests with raw logs or vague descriptions instead of redacted diagnostics and reproduction steps.

Common questions

What makes support evidence useful?

Useful evidence is fresh, reproducible, timestamped, redacted, tied to a specific screen or failure, and routed to the right owner or support channel.

Who should own this decision?

The support lead or site administrator should own the report until it is routed to hosting, provider, CodeCanyon support, or internal operations.

What should stop the rollout?

Stop when evidence is stale, reproduction is missing, redaction is uncertain, or the support route is wrong.

Related operations

• Review Connect an AI Provider.
• Use Incident Response Runbook when production behavior is affected.
• Use Contacting Support before opening a support request.
• Use Support SLA and Escalation Matrix before customer-visible support routing.
• Use Error Reports and Support Codes before redacting or sharing an error report.
• Use Migration and Reindex Review before restarting workflows after updates.
• Use Post-Incident Review and Prevention before returning paused work to normal scope.
• Use Changelog and Release Note Review before release decisions.