Esta página ainda não está disponível no idioma selecionado. Estamos trabalhando para adicionar mais opções de idiomas. Agradecemos sua compreensão.
- What's New
- Product Bulletin
- Service Overview
- Getting Started
-
User Guide
- Edge Computing with IEF
- Service Instances
-
User Guide (Professional)
- Node Management
-
End Device Management
- End Devices and Device Twins
- Device Templates
- End Devices
- Binding an End Device to an Edge Node
- Device Twin Working Principles
- Migrating Device Data to the Cloud
- Performing Security Authentication Using Certificate
-
MQTT Topics
- Device Twin Update
- Device Twin Delta
- Device Member Update
- Device Property Update
- Device Member Acquisition
- Device Member Acquisition Result
- Device Twin Acquisition
- Device Twin Acquisition Result
- Device Twin Modification
- Device Twin Modification Result
- Encryption Data Request
- Encryption Data Acquisition
- Alarm Reporting
- Alarm Clearance
- Custom Topics
- Containerized Application Management
- Edge-Cloud Messages
- Batch Management
- Auditing
- Permissions Management
-
User Guide (Platinum)
- Node Management
-
End Device Management
- End Devices and Device Twins
- Device Templates
- End Devices
- Binding an End Device to an Edge Node
- Device Twin Working Principles
- Migrating Device Data to the Cloud
- Performing Security Authentication Using Certificate
-
MQTT Topics
- Device Twin Update
- Device Twin Delta
- Device Member Update
- Device Property Update
- Device Member Acquisition
- Device Member Acquisition Result
- Device Twin Acquisition
- Device Twin Acquisition Result
- Device Twin Modification
- Device Twin Modification Result
- Encryption Data Request
- Encryption Data Acquisition
- Alarm Reporting
- Alarm Clearance
- Custom Topics
- Containerized Application Management
- Application Mesh
- Edge-Cloud Messages
- Batch Management
- Auditing
- Permissions Management
- Best Practices
-
API Reference
- Before You Start
- API Overview
- Calling APIs
-
API
-
Edge Node Management
- Registering an Edge Node
- Updating an Edge Node
- Querying Details About an Edge Node
- Querying a List of Edge Nodes
- Deleting an Edge Node
- Starting or Stopping an Edge Node
- Updating End Devices for an Edge Node
- Querying Node Certificates
- Creating a Node Certificate
- Deleting a Node Certificate
- Performing an Edge Node Upgrade Check
- Upgrading Edge Nodes
- End Device Management
- End Device Template Management
-
Application Template Management
- Creating an Application Template
- Updating an Application Template
- Querying Details About an Application Template
- Querying a List of Application Templates
- Deleting an Application Template
- Creating an Application Template Version
- Updating an Application Template Version
- Querying Details About an Application Template Version
- Querying a List of Application Template Versions
- Deleting an Application Version
- Deployment Management
- Tag Management
- ConfigMap Management
- Secret Management
- Endpoint Management
- Rule Management
- System Subscription Event Management
- Batch Node Management
- Batch Job Management
- Quota Management
- Service Management
-
Edge Node Management
- Data Structure
- Permissions Policies and Supported Actions
- Appendix
- Change History
- SDK Reference
-
FAQs
-
Edge Node FAQs
- What Do I Do If an Edge Node Is Faulty?
- What Are the Fault Scenarios of Edge Nodes?
- What Do I Do If Edge Node Management Fails?
- How Do I Obtain the Latest Device Properties After Device Properties Are Updated?
- What Operations Can I Perform After a Device Is Associated with a Node?
- Does an Edge Node Support Multiple GPUs?
- Can I Change the GPU on a Running Edge Node?
- How Is Edge Environment Security Protected?
- Why Cannot I View Monitoring Data on an Edge Node?
- How Can I Restore a Deleted Edge Node?
- What Is the Impact of IP Address Changing on an Edge Node?
- What Do I Do If the NTP Configuration Cannot Be Modified?
- How Do I Synchronize Time with the NTP Server?
- How Do I Obtain the IP Addresses of IEF Cloud Services?
- What Do I Do If the Edge Node Space Is Insufficient?
- How Do I Set Docker Cgroup Driver After Installing Docker on an Edge Node?
-
Edge Application FAQs
- What Do I Do If an Application Fails to Be Delivered to an Edge Node?
- What Do I Do If a Containerized Application Fails to Be Started on an Edge Node?
- What Do I Do If a Containerized Application Fails to Be Upgraded?
- What Do I Do If a Container Image Fails to Be Pulled?
- Why Cannot I View Application Logs and System Logs?
- How Do Applications Schedule GPU Resources?
- How Do I Control the Disk Space Occupied by a Container Engine?
- What Do I Do If a Containerized Application Cannot Access External IP Addresses
- What Do I Do If the Ascend AI Accelerator Card (NPU) Is Abnormal?
-
Edge-Cloud Message FAQs
- What Is Route Management?
- What Is a Message Endpoint in Route Management?
- What Is a Route?
- Why Does a Route Fail to Be Created?
- What Can I Do If a Message Fails to Be Forwarded over a Route?
- What Is the Impact of Disabling a Route?
- What Can I Do If SystemEventBus (MQTT Broker) of an Edge Node Fails to Be Connected?
- Network Management FAQs
- Basic Concept FAQs
-
Others
- Region and AZ
- What Are the Specifications of Edge Nodes Supported by IEF?
- What Are the Differences Between Device Properties and Device Twins?
- What Programming Language Is Required for IEF Development?
- Do I Need to Prepare Edge Nodes by Myself?
- Can I Still Use the Previously Delivered Applications After My Account Is in Arrears?
- What Are the Differences Between IEF and IoT Edge?
- What Do I Do If an Agency Fails to Be Automatically Created?
- How Can I Deal With Insufficient Permissions?
- How Will the Multi-AZ Reconstruction of SWR Application Container Image Data Affect IEF?
-
Edge Node FAQs
-
Edgectl User Guide
- Overview
- Installing edgectl
- Example: Using edgectl to Run Commands
-
edgectl Commands
- Querying the edgectl Version
- Managing the Product Lifecycle
-
Checking an Item
- Checking the Hardware Architecture
- Checking the CPU
- Checking the Memory
- Checking Hard Disks
- Checking the Domain Name Resolution Function
- Checking Docker
- Checking Network Connectivity
- Checking the Installation Status of the IEF Software
- Checking the Running Status of the IEF Software
- Checking GPUs
- Checking NPUs
- Checking Processes
-
Error Codes
- Error Code Overview
- Common Errors
- Permission Errors
-
OS Errors
- ERROR3001 Failed to obtain the hardware architecture
- ERROR3002 Unsupported hardware architecture
- ERROR3101 Failed to obtain the CPU information of the node
- ERROR3102 Failed to obtain the real-time CPU status of the node
- ERROR3103 Insufficient node CPU
- ERROR3201 Failed to obtain the memory information of the node
- ERROR3202 Insufficient node memory
- ERROR3301 Failed to obtain the disk information of the node
- ERROR3302 Insufficient disk space on the node
- ERROR3401 Failed to resolve the domain name
- ERROR3501 Failed to check the network
- ERROR3601 Failed to obtain the process information of the node
- ERROR3602 Failed to obtain the total number of processes on the node
- ERROR3603 Number of remaining processes on the node is insufficient
-
IEF Software Errors
- ERROR5001 Incomplete IEF software installation
- ERROR5002 IEF software is not completely running
- ERROR5003 Failed to read the IEF software configuration file
- ERROR5004 Docker is not enabled for the IEF software
- ERROR5005 Failed to start the IEF software
- ERROR5006 Failed to stop the IEF software
- ERROR5101 No application data found
- ERROR5102 No status data of the application found
- ERROR5103 Container is not started
- ERROR5104 Container exits abnormally
- ERROR5105 Failed to pull the container image
- ERROR5106 Container exits
- Third-Party Dependency Errors
- Data Collection Errors
- Internal Errors
- Videos
- General Reference
Copied.
What Do I Do If the Ascend AI Accelerator Card (NPU) Is Abnormal?
Symptom
The NPU-enabled application fails to be delivered or cannot run.
Solution
If the NPU-enabled application fails to be created:
An application applying for NPU resources must be deployed on a node with the Ascend AI accelerator card enabled. If you deploy an application that applies for NPU resources on a node without an Ascend AI accelerator card enabled during node registration, the system displays a message indicating that the application fails to be created.
As shown in the following figure, select an Ascend AI accelerator card based on the model when registering an edge node.
For a node with Ascend AI accelerator card enabled, you can view the AI accelerator card information and check the healthy chip list on the node details page.
If the NPU-enabled application running status is abnormal:
- Ensure that the number of Ascend AI accelerator cards applied for by the current application is not greater than that of healthy chips on the node. Otherwise, the application scheduling will fail.
Check the number of Ascend cards on the Upgrade tab page of the containerized application details page:
- Log in to the edge node, and check whether the NPU plug-in is normal.
docker ps -a |grep npu-plugin
- If the container status is abnormal, restart the container.
docker restart $containerID
- If the fault persists, go to What Do I Do If an Application Fails to Be Delivered to an Edge Node? and What Do I Do If a Containerized Application Fails to Be Started on an Edge Node?.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot