Feature Introduction

The eSurfing Cloud CDN acceleration platform supports automated service metric monitoring and alerting functions. Customers can set up relevant monitoring and alerting rules based on actual business needs. When alert rules are triggered, the eSurfing Cloud monitoring system will send alert messages via SMS, email, WeChat Work, DingTalk, etc., as per the customer's configured notification methods, notifying them to intervene promptly and address related issues.

Currently supported common monitoring/alert metrics include but are not limited to:

Bandwidth/Traffic: Upper/lower limit monitoring, sudden increase/decrease monitoring.
Request count: Upper/lower limit monitoring, sudden increase/decrease monitoring.
Status codes: Monitoring abnormal status code (5xx/4xx) counts, abnormal status code (5xx/4xx) ratio.

Applicable Scenarios

If your business involves large file downloads or audio/video on-demand services and frequently encounters bandwidth/traffic fluctuations, you may consider focusing on setting up bandwidth/traffic-related monitoring metrics and appropriate thresholds for monitoring and alerting.
If your business deals with static small files, such as government and enterprise websites, financial securities, e-commerce, and news media sites, and is more concerned about changes in user traffic and QPS, you can consider focusing on setting up request count/QPS-related monitoring metrics and appropriate thresholds for monitoring and alerting.
If your business is sensitive to service availability, you can set up status code-related monitoring metrics and appropriate thresholds to monitor the operational status of the business in real-time, ensuring timely alerting and manual intervention when abnormalities occur.

Configuration Instructions

This feature does not currently support customer self-service configuration. If you need to use it, please call the hotline at (852) 31000000 or send an email to global.noc@chinatelecomglobal.com to contact the after-sales service team for manual operation.

Please provide the following information:

Parameter Name	Description	Default Value	Example
Alert Name	The name of the alert is used to distinguish different alert tasks. When customers receive alert messages, they mainly distinguish and identify different alerts based on the alert name.	None.	Edge bandwidth surge alert.
Monitoring Scope	The scope of the alert rule: All domains or specific domains. 1. All domains: All CDN accelerated domains under your account. 2. Specific domains: Monitor only specific domains, supporting target domains (whitelist) or exception domains (blacklist).	All domains.	All domains.
Monitoring Metrics	Specific alert rules, triggering alerts when monitoring data meets alert conditions. 1. Monitoring period: Trigger alerts when conditions are met during "00:00 to 24:00". Support for multiple periods and multiple monitoring metrics, with the option to select the relationship between different monitoring metrics, for example: any or all. "Any" means that an alert is triggered if any of the metrics reaches the alert threshold; "All" means that an alert is triggered only if all metrics reach the alert threshold simultaneously. 2. Monitoring metrics: See below table for details: Monitoring Metrics Description.	"00:00 to 24:00" period meets "any" conditions.	Example 1: Monitoring period 1: Trigger alerts when conditions are met from 18:00 to 22:00. 1. Edge bandwidth growth rate (%) > 50. 2. 5xx ratio (%) ≥ 2. Example 2: Monitoring period 2: Trigger alerts when conditions are met from 00:00 to 24:00 with "all" conditions. 1. Edge bandwidth growth rate (%) > 100. 2. 5xx ratio (%) ≥ 5.
Trigger Conditions	To avoid alerts being affected by occasional network fluctuations and other factors causing false alarms (alarms that do not need to be processed), the monitoring platform supports setting the following two parameters: 1. Duration: Indicates that an alert is triggered only after an abnormality lasts for a period of time. Supports options of 300 seconds and 1500 seconds. 2. Consecutive Trigger Times: Within the duration, the alert is triggered only after the alert condition is met for a specified number of consecutive times.	1. Duration: Default 5 minutes. 2. Consecutive Trigger Times: Default 3 times.	1. Duration: 10 minutes. 2. Consecutive Trigger Times: 5 times.
Alert Silence Period	After an alert occurs and does not return to normal, how long to wait before sending another alert notification, in minutes. If a monitoring metric exceeds the alert threshold and remains above the threshold during the silence period, no alert notification will be sent during the silence period. If the monitoring metric does not return to normal after the silence period, the monitoring will send another alert notification. For example: When the alert silence period is set to 60 minutes, if the alert does not return to normal, the monitoring will send another alert notification after 60 minutes.	5 minutes.	Silence Period: 10 minutes.
Alert Channels	Notification channels when alerts are triggered. Supports voice service, SMS, email, WeChat Work (or WeChat Work robots), and DingTalk robots.	None.	SMS.
Alert Contact Group	The contact group to which the alert is sent. An alert contact group is a group of alert contacts, which can include one or more alert contacts.	None.	For example: Zhang San Phone Number 1. Li Si Phone Number 2.
Alert Content	If there are no special requirements, this item can be ignored. The system will automatically generate alert messages based on specific metric templates.

Precautions

When configuring multiple alert metrics simultaneously, the platform supports alerting when "any" or "all" conditions are met. When submitting a ticket, please specify clearly.
If you are not very familiar with the characteristics of your own business, you can first submit a threshold scheme based on experience. Your business operations colleagues will help you make some basic judgments and suggestions, and you can try it first. Then, based on the actual operation situation online, you can submit tickets for optimization later, until you find suitable monitoring metrics and thresholds.

Alert Example

Example of Alert Requirement Submission:

Alert Name: Bandwidth Surge and Availability Alert.
Monitoring Scope: www.example.com (single domain).
Monitoring Metrics: Trigger alerts when conditions are met from 18:00 to 22:00 with "any" conditions.

Condition 1: Edge bandwidth growth rate (%) > 50.
Condition 2: 5xx ratio (%) ≥ 2.

Trigger Conditions:

Duration: 10 minutes.
Consecutive Trigger Times: 5 times.

Alert Silence Period: 30 minutes.
Alert Channels: WeChat Work robot.
Alert Contacts: Wang Wu 186XXXXXXXX.
Alert Content: System default.

WeChat Work Alert Message Example:

[alert]-www.example.com:-Bandwidth Surge and Availability Alert:

Wang Wu.
Customer Email: XXXX.
domain=[www.example.com].

Alert Notification ID=XXXXXXXXXXXXX-XXXXXXXX.
[Bandwidth Surge and Availability Alert] occurred - Customer Level.
Current value is [5xx ratio (%) = 3].
Trigger condition [5xx ratio (%) exceeds 2].
Fault duration XXX minutes.
Start time: 2023-11-20 06:25:01.
Last alert time: 2023-11-20 11:30:04.
During this period, X alerts occurred and X recovered.

Content Delivery Network

Montoring and Alerts