Skip to main content

Diagnose by failure class

Use this page when you already know something is wrong and need the fastest public recovery route.

Signature and encryption failures

  1. Verify signer identity recovery matches trader.
  2. Verify payload bytes are unchanged between signing and encryption.
  3. Verify operator encryption key fetch succeeds before submission.
  4. Recheck Request Signing Reference and Request Encryption Reference before retrying.

Safety failures

  1. Map safety-failure variant from response.
  2. Check tick-size, minimum-size, notional, market support, and collateral constraints.
  3. Rebuild, resign, and re-encrypt request.
  4. Use Trading Safeties and Guards and Margin Requirements when the failure is really a risk-constraint issue rather than malformed input.

Sequencing versus execution confusion

  1. Treat receipt as sequencing acknowledgment only.
  2. Track ORDER_UPDATE and account updates for outcome.
  3. Reconcile with REST snapshot after reconnect.

Realtime reconnect instability

  1. Reconnect WebSocket.
  2. Resubscribe feeds with explicit filters.
  3. Re-bootstrap state from REST.
  4. Recheck Realtime Connection Model if the issue repeats.

Escalate by symptom

SymptomBest next route
bad signature, bad nonce, or request-rejection uncertaintyHow to Handle API Errors and Recovery and Error Reference
lifecycle mismatch after receiptHow to Handle Order Sequencing vs Execution and Order and Account Event Reference
repeated docs mismatch or unresolved public ambiguitySupport Channels
KYC or Blockpass-specific failureSupport Channels and then Blockpass support if the public docs do not resolve it

References

Boundary rule

This page is for public recovery triage. It is not a private operator incident runbook and it should not be used to infer unpublished restricted-surface procedures.
Last modified on April 13, 2026