監視自訂管道 monitor-custom-channel

進一步瞭解自訂管道量度

• 成功的呼叫：傳回有效回應且沒有錯誤的HTTP呼叫總數。

• 4xx/5xx錯誤：由於使用者端(4xx)或伺服器端(5xx)錯誤，強調設定問題或端點失敗而失敗的呼叫數目。

• 逾時呼叫：因超過最大回應時間而失敗的呼叫數。 這有助於顯示外部端點的延遲或效能問題。

• 預先呼叫失敗：對外部端點進行HTTP呼叫之前失敗的自訂通道傳送數目。 這些失敗發生在Journey Optimizer自己的基礎結構層 — 而不是外部系統中。 共有三種類別：

  table 0-row-2 1-row-2 2-row-2 3-row-2
  類別	說明
  驗證失敗 (AUTH_*)	Journey Optimizer無法取得或重新整理OAuth權杖或呼叫端點所需的認證。 檢查連結到通道設定的API認證是否有效且尚未過期。
  要求產生錯誤 (REQUEST_GENERATION_ERROR)	Journey Optimizer無法建構有效的HTTP要求 — 例如，因為無法解析URL範本或缺少必要的個人化欄位。
  HTTP剖析錯誤 (HTTP_PARSE_ERROR)	Journey Optimizer從端點收到回應，但無法將其剖析為可用的結構。

  note tip
  TIP
  呼叫前失敗表示Journey Optimizer端或通道設定有問題，而不是您的外部端點有問題。 檢閱您的API憑證和必要的裝載欄位，開始疑難排解。

• 平均延遲：所有HTTP呼叫的平均端對端回應時間（以毫秒為單位），包括成功的呼叫、錯誤和逾時。

AI Knowledge Reference

This section contains structured knowledge intended to support interpretation, retrieval, and question answering related to this topic.

For complete understanding, this information should be combined with the documentation on this page. Neither source is intended to stand alone; the page describes the feature, while this section provides additional context that helps disambiguate terminology, intent, applicability, and constraints.

• TL;DR: This page explains how to monitor custom channel delivery performance in Journey Optimizer using out-of-the-box campaign and journey reports and the dedicated custom channel monitoring dashboard, and how to troubleshoot common delivery issues.

Intents:

• Monitor custom channel delivery performance through out-of-the-box campaign and journey reports
• Access and use the custom channel monitoring dashboard
• Interpret custom channel metrics such as successful calls, errors, timeouts, and pre-call failures
• Analyze outcomes and latency over time and break them down by endpoint, channel, campaign, and journey
• Troubleshoot common custom channel delivery symptoms and apply recommended resolutions

Glossary:

• Custom channels monitoring dashboard: A dedicated dashboard accessed from Administration > Channels > Channel Builder > Custom channels monitoring that monitors the reliability and performance of the API calls Journey Optimizer makes to external endpoints (product-specific)
• Pre-call failures: Custom channel sends that failed before the HTTP call was ever made to the external endpoint; these occur in Journey Optimizer’s own infrastructure layer and fall into three categories: authentication failures (AUTH_*), request generation errors (REQUEST_GENERATION_ERROR), and HTTP parse errors (HTTP_PARSE_ERROR) (product-specific)
• Timeout calls: Number of calls that failed because they exceeded the maximum response time (product-specific)
• Insight Builder: A tool used to create custom visualizations and dashboards based on custom channel metrics (product-specific)
• Successful deliveries: Messages for which the endpoint returned an HTTP 2xx response (product-specific)

Guardrails:

• Clicks metric tracking requires a subdomain delegated for custom channels.
• Pre-call failures indicate a problem on the Journey Optimizer side or in the channel configuration, rather than an issue with your external endpoint; start troubleshooting by reviewing API credentials and required payload fields.
• Average latency is the average end-to-end response time in milliseconds for all HTTP calls, including successful calls, errors, and timeouts.
• Report granularity depends on the selected time range: a 7-day report shows KPIs per day, a 1-day range shows KPIs per hour, and a 1-hour range shows KPIs per minute.
• HTTP 401/403 errors indicate an authentication failure; update the credentials in Administration > Channels > API credentials.
• HTTP 429 errors indicate the external endpoint is throttling requests; review your endpoint’s rate limits and reduce the throttling setting in the Channel Builder policy configuration.

Terminology:

• Canonical name: Custom channels monitoring dashboard — Acronym: n/a — variants: custom channel monitoring dashboard, Custom channel metrics
• Synonyms: “Attempted deliveries” = total number of messages sent to the external endpoint
• Do not confuse: “Successful deliveries” (report metric: messages for which the endpoint returned an HTTP 2xx response) ≠ “Successful calls” (monitoring metric: total HTTP calls that returned a valid response without error)
• Do not confuse: “4xx/5xx errors” (failed calls due to client-side or server-side errors) ≠ “Timeout calls” (calls that exceeded the maximum response time) ≠ “Pre-call failures” (sends that failed before the HTTP call was made)

FAQ:

• Q: Where can I monitor custom channel performance? — Through out-of-the-box campaign and journey reports and the dedicated custom channel monitoring dashboard, accessed from Administration > Channels > Channel Builder > Custom channels monitoring.
• Q: What does the Clicks metric require? — The Clicks metric counts link clicks tracked in the payload and requires a subdomain delegated for custom channels.
• Q: What are pre-call failures? — Custom channel sends that failed before the HTTP call was ever made to the external endpoint; they occur in Journey Optimizer’s own infrastructure layer and include authentication failures, request generation errors, and HTTP parse errors.
• Q: My calls return HTTP 429 errors. What should I do? — The external endpoint is throttling requests from Journey Optimizer; review your endpoint’s rate limits and reduce the throttling setting in the Channel Builder policy configuration.
• Q: How does the report time granularity change with the time range? — A 7-day report shows one data point per day, a 1-day range shows KPIs per hour, and a 1-hour range shows KPIs per minute.
• Q: How do I troubleshoot unresolved personalization tokens? — Verify the XDM attribute path is correct and add a default value fallback, for example {{profile.person.name.firstName \| default("Valued Customer")}}.