How to Monitor FunctionGraph in Huawei Cloud
Site24x7 brings deep observability to your HuaweiCloud FunctionGraph environment, giving your team real-time visibility into invocation throughput, error and throttle rates, concurrent instance scaling, execution latency, and resource consumption. This ensures your serverless workloads stay fast, efficient, and always on.
Use cases
Error triage: Separate Error Count into System Error Count and Function Error Count to distinguish infrastructure failures from code issues.
Throttle prevention: Throttle Count signals concurrency limits have been breached, causing requests to be rejected. Track alongside Concurrency and Instance Usage to raise limits or reserve instances before throttling cascades to dependent services.
Latency protection: Average Duration masks tail latency issues. Track Duration P95 and P99.9 alongside Maximum Duration to catch outlier invocations breaching latency SLOs before they accumulate and impact user-facing response times.
Setup and configuration
FunctionGraph resources are auto-discovered and monitored during the Huawei Cloud integration. To enable monitoring, follow the steps below:
- Navigate to Cloud > Huawei > Add Huawei Monitor. Follow the steps to add a Huawei Cloud monitor.
- While adding or editing a Huawei Cloud monitor, select Function Graph from the Service/Resource Types drop-down and click Save.
- Navigate to Cloud > Huawei, select the created Huawei monitor, and click FunctionGraph.
Supported metrics
Invocations
Metric name | Description | Units |
| Invocation Count | The total number of times the function has been invoked within the monitoring period. | Count |
| Error Count | The total number of failed invocations across all error types. | Count |
| Error Rate | The percentage of function invocations that resulted in an error. | Percentage |
| Throttle Count | The number of invocation requests rejected due to the concurrent execution limit being exceeded. | Count |
| Concurrency | The number of function requests being executed concurrently at a given point in time. | Count |
| System Error Count | The number of invocation failures caused by infrastructure or platform-level errors. | Count |
| Function Error Count | The number of invocation failures caused by errors within the function code itself. | Count |
Instances
Metric name | Description | Units |
| Reserved Instances | The number of pre-provisioned instances allocated to the function to eliminate cold starts. | Count |
| Pay Per Use Instances | The number of on-demand instances currently running and billed per invocation. | Count |
| Running Instances | The total number of function instances currently active and handling requests. | Count |
| Instance Usage | The percentage of available function instances currently in use. | Percentage |
Latency
Metric name | Description | Units |
| Average Duration | The average execution time of function invocations within the monitoring period. | Millisecond |
| Maximum Duration | The longest execution time recorded for a single function invocation. | Millisecond |
| Minimum Duration | The shortest execution time recorded for a single function invocation. | Millisecond |
| Duration P50 | The execution time below which 50% of function invocations are completed. | Millisecond |
| Duration P80 | The execution time below which 80% of function invocations are completed. | Millisecond |
| Duration P95 | The execution time below which 95% of function invocations are completed. | Millisecond |
| Duration P99 | The execution time below which 99% of function invocations completed. | Millisecond |
| Duration P99.9 | The execution time below which 99.9% of function invocations completed. | Millisecond |
Resources
Metric name | Description | Units |
| Memory Used | The average amount of memory consumed per function invocation. | MB |
| Function Cost | The compute resource consumption of the function, expressed in MB-seconds. | MB |
Threshold configuration
You can configure thresholds and alerts for all FunctionGraph metrics to detect performance degradation proactively or connection issues.
- Go to Admin > Configuration Profiles > Threshold and Availability.
- Create or edit your Threshold Profile for Function Graph.
- Assign the profile to the respective monitors to trigger alerts.
IT Automation
Use Site24x7's IT Automation to resolve common issues with FunctionGraph performance:
- Go to Admin >IT Automation Templates. Then, click Add Automation Templates.
- Create an automation rule by selecting the automation Type (e.g., Server reboot, clear queue).
- Map the created rules to the FunctionGraph, for automatic execution during alerts.
Configuration rules
Use Configuration Rules to simplify bulk setup across FunctionGraph instances. Automatically assign Threshold Profiles, Notification Profiles, Tags, and Monitor Groups when new monitors are discovered.