OSS monitoring provides system metrics for performance and running status, plus a custom alert service to track requests, analyze usage, collect statistics on business trends, and diagnose issues.
Metrics of the OSS monitoring service are classified into basic service metrics, performance metrics, and billable usage metrics. For more information, see Monitoring metrics reference.
Real-time metrics
Real-time metrics expose traffic fluctuations to help you analyze business scenarios. Metrics (except billable usage metrics) are collected at minute-level granularity with sub-minute output delay. Data collected each minute is aggregated into a single value and displayed within one minute.
Billable usage metrics
Billable usage metrics follow these collection rules:
-
Data is collected hourly. Each hour's metering data is aggregated into a single value.
-
Data is displayed with a delay of nearly 30 minutes.
-
The collection time is the start of each statistical period.
-
The collection deadline is the end of the last statistical period in the current month. If no data is produced, the deadline defaults to 00:00:00 on the first day of the month.
-
A maximum amount of metric data is pushed for display. For precise data, go to the Billing Management console and click Usage Records.
Example: If you call PutObject 10 times per minute, the PUT request count for 08:00:00–09:00:00 on May 10, 2018 is 600 (10 × 60). The collection time is 08:00:00, and data appears around 09:30:00. If this is the last data since 00:00 on May 1, 2018, the collection deadline is 09:00:00 on May 10. If no data is produced in May 2018, the deadline is 00:00:00 on May 1, 2018.
Alert service
Each account supports up to 1,000 alert rules. Rules can be configured for monitoring metrics in addition to billable usage and statistical metrics. Multiple rules can apply to a single metric.
-
For more information about the metrics of the OSS monitoring service, see Monitoring metrics reference.
Data retention
Metric data is retained for 31 days and automatically cleared upon expiration. To store or analyze older data, read from CloudMonitor storage. Access metric data by using the CloudMonitor API.
The OSS console displays the last seven days of metric data. For older data, use CloudMonitor SDKs. Access metric data by using the CloudMonitor OpenAPI.
Access metrics through CloudMonitor OpenAPI
Use the CloudMonitor API to access OSS metric data:
Monitoring, diagnosis, and troubleshooting
Service monitoring, diagnosis, and troubleshooting covers:
-
Track system status and performance with OSS monitoring.
-
Use monitoring and logging to diagnose problems and correlate log data.
-
Common problems and troubleshooting methods.
Considerations
Bucket names must be globally unique. If you recreate a bucket with the same name as a deleted bucket, the previous monitoring and alert rules apply to the new bucket.