FAQ ​

1. Newly added package not displayed normally after adding package in package management page ​

It may be caused by synchronization failure due to network fluctuations, etc. Manually click "Sync" to resolve.

2. Error adding remote repository: unable to find index.yaml, please provide correct ChartMuseum project url ​

The remote repository link is filled incorrectly. Need to ensure that the url specifies the specific project name in the harbor repository. For example, openFuyao official repository https://helm.openfuyao.cn specifies project "openfuyao-catalog".

3. After inviting platform administrator to join cluster, corresponding permissions do not take effect after setting cluster role for them ​

If the user is already in platform admin role, after setting cluster-level role for them, permissions are still platform admin permissions. Platform admin role has the highest permission priority.

4. cluster admin user cannot invite other users to join cluster ​

The main reason is that this user does not have platform admin permission. Need to configure platform admin role for this user first before inviting other users.

5. Deployment fails when deploying again after application uninstallation ​

Application uninstallation requires certain time. Immediately clicking deployment may cause deployment failure due to incomplete uninstallation. It is recommended to wait a few minutes before deploying again.

6. Abnormality occurs when accessing openFuyao platform log module using Firefox browser ​

When using Firefox browser to access openFuyao platform log module, if page cannot be opened or data display is abnormal, please manually clear browser cache.

7. Time difference between browser and server may affect some monitoring data ​

Current time difference between browser and server may cause data delay or misalignment. To avoid continuous impact, it is recommended to check local device time settings and ensure synchronization with server time. After adjusting browser time, changes may take some time to take effect.

8. How to modify openFuyao management interface exposed port ​

Current openFuyao management interface default port number is 31616. If you want to modify default port number to other port, for v25.03 version please follow these steps.

Note:
This step uses installation package fuyao-v25.03-openEuler-22.03-amd64 as an example. Other installation package processing steps are similar.

• Operation steps

  1. After downloading openFuyao installation package, extract shell script compressed package under asset directory in installation package.

     shell

     cd asset
     tar -zxvf fuyao-openEuler-22.03-amd64.tar.gz

  2. Edit fuyao-openEuler-22.03-amd64/ingress/ingress-nginx/resource/ingress-nginx.yaml file, find nodePort: 31616 around line 372, modify 31616 to other port number. The modification location is as follows.

     shell

     367 - appProtocol: https
     368   name: https
     369   port: 443
     370   protocol: TCP
     371   targetPort: https
     372   nodePort: 31616 # modification item

  3. After completing modification, delete original fuyao-openEuler-22.03-amd64.tar.gz, and compress fuyao-openEuler-22.03-amd64 to fuyao-openEuler-22.03-amd64.tar.gz.

     shell

     rm -f fuyao-openEuler-22.03-amd64.tar.gz
     tar -czvf fuyao-openEuler-22.03-amd64.tar.gz ./fuyao-openEuler-22.03-amd64/*

  4. At this point, port number modification is completed. You can continue to install cluster according to Installation Guide. After completing cluster installation, need to use modified port number to access openFuyao management interface.

9. Pod startup fails, kubectl logs shows failed to create fsnotify watcher: too many open files ​

It may be because system default fs.inotify.max_user_instances is too small. Reset this value on the node where pod is located. If it is multi-node cluster installation and deployment, can reset fs.inotify.max_user_instances for all nodes to avoid errors.

Solution 1: Temporarily set fs.inotify.max_user_instances value.

sudo sysctl fs.inotify.max_user_instances=8192

Note: Need to set again when re-entering console.

Solution 2: Permanently save fs.inotify.max_user_instances value.

echo fs.inotify.max_user_instances=8192| tee -a /etc/sysctl.conf && sudo sysctl -p

10. Insufficient space under root partition but other partitions still have remaining space ​

For the situation where root partition space is insufficient but other partitions have surplus space, you can consider migrating container-related directories to other partitions. For example, if a partition with remaining space is mounted at /home, you can perform migration according to the following steps:

1. Create directory under /home.

mkdir /home/containerd

2. Stop kubelet and containerd in order.

systemctl stop kubelet
systemctl stop containerd

3. Modify containerd configuration file.
   First find the configuration file used by containerd, default is /etc/containerd/config.toml. Edit file, record the original root directory, modify to root = "/home/containerd".

4. Migrate original files to newly created folder.

cp -r ${original_root}/* /home/containerd/

5. Restart containerd and kubelet in order again.

systemctl start containerd
systemctl start kubelet

Note:
The feasibility and stability of this solution in production scenarios have not been verified. Not recommended for use in production environment.