All Products
Search

This Product

    Elastic Desktop Service:FAQ for administrators managing cloud computers

    Document Center

    Elastic Desktop Service:FAQ for administrators managing cloud computers

    Last Updated:Dec 08, 2025

    This topic provides answers to some frequently asked questions (FAQ) about managing cloud computers in Elastic Desktop Service (EDS) Enterprise as administrators.

    Table of contents

    Category

    Question

    Creation and deployment

    User and assignment

    Query and management

    Storage and networking

    Cloud computer connections

    Application and peripheral

    Cloud computer startup, restart, and wakeup errors

    Extension and others

    How can I troubleshoot EDS API call errors?

    Creation and deployment

    What should I do if cloud computers fail to be created?

    Problem description

    When I created cloud computers, the EDS console displayed a creation failure.

    Cause

    The queuing mechanism is activated when an excessive number of administrators concurrently attempt to create cloud computers. If the process is not finalized within one hour, the EDS console will indicate a failed creation.

    Solution

    You can release the cloud computers that failed to be created and then recreate them. If the failed cloud computers are not released within 7 days, they will be automatically reclaimed. No billing will be incurred for cloud computers that failed to be created.

    What should I do if a cloud computer remains in the Registering state?

    Problem description

    The cloud computer is stuck in the "Registering" state.

    Solution

    Can I migrate a cloud computer from the China (Hangzhou) region to the China (Beijing) region?

    No. Cross-region migration of cloud computers is not supported.

    How do I quickly duplicate a cloud computer?

    If you need to quickly create a copy of a cloud computer, you can follow these steps:

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Resources > Cloud Computers.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Cloud Computers page, find the target cloud computer. In the Actions column, click the ⋮ icon and select Create Image. This will generate a custom image of the cloud computer for future use. For more information, see Create an image.

    5. After creating a custom image, create a custom cloud computer template based on that custom image. For more information, see Create and manage custom templates.

    6. Once the custom template is created, create a cloud computer based on this custom template. The newly created cloud computer will be identical to the one used to make the custom image, thereby achieving the goal of quickly replicating the cloud computer. For more information, see Create cloud computers.

    What do I do if I do not receive the email containing my account information after I paid for EDS Enterprise Edition?

    After you have paid for EDS Enterprise Edition, EDS will send an email containing your account information to the email address you provided during purchase. If you have completed your payment but have not received the email, perform the following checks:

    • Double check your email address: Make sure the email box you are checking is the email address you provided during purchase. If you find out that you specified a wrong email address during purchase, modify it promptly.

    • Check the trash folder: The email may be mistakenly considered a trash or advertisement email.

    • Wait for another period of time: In some cases, it takes 5 to 15 minutes for the email to be sent. Please wait for some more time and check your inbox again.

    If the email is still absent, submit a ticket to contact Alibaba Cloud technical support.

    User and assignment

    Why are there no users available to select when I create cloud computers?

    If no users appear in the list when you select authorized users, it means you have not yet created any users. You can create users based on the user account system type. For more information, see Create a convenience account or Create and manage enterprise AD accounts.

    Is it possible for multiple end users to connect to and use a single cloud computer at the same time?

    • One cloud computer can be assigned to multiple end users, but only one end user can connect to and use it at any given time. Until the currently connected end user disconnects from the cloud computer, no other end users can connect to it. For more information, see Assign cloud computers to users.

      Note

      Each time a new user logs on to a cloud computer, the session with the previous user is forcibly logged off, closing all windows and applications they opened. However, any user data written to the disk is retained.

    • If needed, you can set up a one-to-many share, configure the number of concurrent connections, and assign it to multiple end users. Each end user can connect to this one-to-many share at the same time, but they will operate in isolated environments.

      Note

      The feature is in invitational preview. If you want to use this feature, submit a ticket.

    What should I do if end users find no cloud computers available after logging onto their clients?

    Follow the steps below to confirm whether cloud computers have been assigned to end users:

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Resources > Cloud Computers.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. From the search criteria list, select Username, enter an end user's name, and press Enter.

      • If the end user already has a cloud computer, ask the user to verify whether the office network ID (formerly workspace ID) or organization ID entered during logon is correct.

      • If the end user has no cloud computer, you can assign one to the user. For more information, see Manage convenience accounts.

    What should I do if end users are unable to create files and folders in the root directory of the C drive?

    Problem description

    When end users use Windows cloud computers within a many-to-many share (formerly known as the cloud computer pool), they already have local admin privileges. However, when attempting to create a folder in the root directory of the C: drive, they still receive an Access Denied error for the target folder.

    Cause

    For cloud computers from a many-to-many share, the system imposes default restrictions on the root directory of the C: drive. Users are prohibited from creating any files or folders in the C: drive's root directory.

    Solution

    To remove the restrictions, follow these steps:

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Resources > Shared Cloud Computer.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Share page, locate the target many-to-many share using the share ID (previously known as the cloud computer pool ID) provided by the end user.

    5. Click the share ID, and select the Command Execution Details tab.

    6. Under the Command Execution Details tab, click Send Remote Commands.

    7. In the Send Remote Commands panel, select the command type and the target cloud computer.

      Note

      For Windows cloud computers, choose either PowerShell or Bat as the command type.

    8. In the Command Content text box, enter the following command and click Execute to remove the restrictions on the root directory of the C: drive: 

      cmd.exe /c "icacls C:\ /grant CDriveLimited:(AD)"

      If the command output displays a message similar to the following, the restrictions on the root directory of the C: drive have been successfully removed: 

      Processed file: C:\
Successfully processed 1 file; Failed to process 0 files.

    End users can now create files and folders in the root directory of the C: drive.

    Query and management

    How do I restart a cloud computer by running a remote command?

    End users may encounter errors such as 22, 5100, 5102, 5202, ConnectTicket.Timeout, UnavailableDesktop.ConnectionBroken, or other abnormal situations (e.g., keyboard unresponsiveness) when connecting to Windows cloud computers. In such cases, restarting cloud computers via a remote command may resolve the issue.

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Resources > Cloud Computers.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Cloud Computers page, locate the target cloud computer with a Running status. In the Actions column, click the ⋮ icon and select Send Remote Commands.

    5. On the Send Remote Commands page, select PowerShell as the command type.

    6. In the Command Content text box, enter the following command and click Execute to restart the target cloud computer: 

      taskkill /im spmonitor.exe -f | taskkill /im aspsvdi.exe -f | taskkill /im edsagent.exe -f

    If the command output displays messages similar to the following, the cloud computer has been successfully restarted: 

    SUCCESS: The process "aspsvdi.exe" with PID 4164 has been terminated.

    The example above uses a Windows cloud computer. Please adjust the actual steps based on your specific environment. For more information about sending remote commands, see Send remote commands.

    How do I upgrade the GRID driver of an Enterprise Graphics cloud computer?

    Prerequisites

    • Before upgrading the GRID driver, ensure that the cloud computer's system disk has at least 4 GiB of available space.

    • A system disk snapshot has been created prior to upgrading. If the upgrade fails, you can restore the system disk data using the snapshot. For more information, see Use snapshots (public preview).

    Procedure

    Windows cloud computers (admin operations)

    For Windows cloud computers, simply upgrade the image to version 1.8.0 or later to complete the GRID driver upgrade. Follow these steps as an administrator to upgrade the image:

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Images Center > Image Updates.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Image Updates page, locate the current image version of the cloud computer. In the Actions column, click Query Involved Cloud Computer and Update Progress.

    5. In the Query Involved Cloud Computer and Update Progress panel, perform one of the following operations as needed:

      • Single operation: Click Update in the Actions column for the target cloud computer. Choose Update Now or Schedule Update. Then, follow the on-screen instructions to proceed.

      • Batch operation: Select multiple cloud computers that require an image upgrade. Click Immediate/Schedule Update at the bottom. Then, follow the on-screen instructions to proceed.

    For more information, see Update an image.

    Windows cloud computers (end user operations)

    For Windows cloud computers, simply upgrade the image to version 1.8.0 or later to complete the GRID driver upgrade. End users can follow these steps to upgrade the image:

    Important

    • Save all opening files before the upgrade.

    • It takes 10 to 15 minutes for the upgrade to complete. You cannot use the cloud computer during the process.

    1. Click Upgrade in the target cloud computer card.

    2. In the dialog box that appears, choose an upgrade time:

      • To upgrade immediately, choose Upgrade Now and select a Status after Upgrade for the cloud computer, either Remain in Current Status, Shut Down, Running, or Hibernate.

      • To upgrade later, choose Scheduled Upgrade and select a start time, either 4 hours, 8 hours, or 12 hours, from the Select Start Time drop-down list.

    3. Click Confirm.

    4. In the Upgraded dialog box, click Got it.

      • If you cannot connect to the cloud computer after an upgrade or need to roll back the upgrade, click Manage on the cloud computer card and choose Restore Points > System Restore Point. Click Restore, select the target system disk or data disk, and click Confirm. Alternatively, contact your enterprise IT administrator for a rollback.

      • If the upgrade fails, please try again or contact your enterprise IT administrator.

    Linux cloud computers (administrator operations)

    For Linux cloud computers, administrators must update the GRID driver via remote commands.

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Resources > Cloud Computers.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Cloud Computers page, restart the target cloud computer.

    5. Open the Send Remote Commands dialog using one of the following methods:

      • Click the cloud computer ID, select the Command Execution Details tab, and then click Send Remote Commands.

      • Click the ⋮ icon in the Actions column and select Send Remote Commands.

    6. In the Command Content text box, execute the following commands to upgrade the GRID driver:

      Important

      • Ensure the cloud computer is disconnected before sending commands.

      • Command execution takes about 2 minutes. 

      if acs-plugin-manager --list --local | grep grid_driver_install > /dev/null 2>&1
then
acs-plugin-manager --remove --plugin grid_driver_install
fi
acs-plugin-manager --exec --plugin grid_driver_install

      Check the command status at the bottom of the output or on the Command Execution Details tab. If it shows Succeeded, the command is successfully executed.

    7. Restart the cloud computer.

    Verification

    • Linux

      Connect to the cloud computer and open NVIDIA X Server Settings. Click X Server Information in the left panel. If NVIDIA Driver Version shows 470.161.03, the GRID driver upgrade succeeded.

    • Windows

      Connect to the cloud computer, right-click the desktop, and select NVIDIA Control Panel. If the version is 474.04, the GRID driver upgrade succeeded.

    Note

    If the upgrade fails, submit a ticket for technical assistance.

    Storage and networking

    What should I do if the cloud computer's disk space does not change after expansion?

    If your cloud computer runs out of disk space, you can upgrade its configuration to meet your business needs. After increasing the disk space, if the available space remains the same, you may need to manually extend the disk partition. For more information, see Scale up the partition of a data disk.

    What should I do if the office network cannot be selected when creating an NAS storage?

    Problem description

    In EDS Enterprise, file sharing between cloud computers within the same office network requires creating an NAS storage. However, the Office Network option is unavailable during NAS creation.

    Cause

    • NAS is not activated.

    • An advanced office network is not created in the target region.

    Solution

    1. Log on to the NAS console and check whether NAS is activated. If NAS is not activated, activate NAS as prompted.

    2. Log on to the Elastic Desktop Service Enterprise console.

    3. In the left-side navigation pane, choose Networks & Storage > Office Networks.

    4. On the Office Network page, check whether an advanced office network is created. If no advanced office network is created, create one. For more information, see Create and manage convenience office networks.

    5. After you perform the preceding steps, create an NAS storage again. For more information, see Mount an NAS file system on a Windows cloud computer.

    What should I do when the error message "The charge type of network package in office site is not allowed to delete" appears while attempting to delete an office network?

    Problem description

    If you no longer need a specific office network, you can release all cloud computers associated with it and then delete the office network. However, when attempting to delete the office network, you may encounter the following error message: The charge type of network package in office site is not allowed to delete.

    Cause

    The office network still has an unbound subscription premium bandwidth plan.

    Solution

    To unbind the subscription premium bandwidth plan from the office network, follow these steps:

    1. Log on to the Elastic Desktop Service Enterprise console.

    2. In the left-side navigation pane, choose Networks & Storage > Premium Bandwidth Plans.

    3. In the upper-left corner of the top navigation bar, select a region.

    4. On the Premium Bandwidth Plan page, find the target premium bandwidth plan and click Disassociate Office Network in the Actions column.

    5. In the Confirm Office Network Disassociation message, click Confirm.

      If the plan changes to the Not Associated state, the plan is unbound from the office network.

    6. Retry deleting the office network. For more information, see Create and manage convenience office networks.

    Cloud computer connections

    What should I do if the "Exceptions occur in the EDS service. Failed to connect to the cloud computer. [22]" message appears?

    Causes

    • The AF_UNIX dependency file required for asp-server connections is deleted. The 1.5.x and 1.6.x images of the file are stored in the C:\windows\Temp directory, which can be accidentally deleted by cleanup tools.

    • The AF_UNIX dependency file is configured with System and Hidden attributes in the 1.7.x image. Then, exceptions may occur when the system connects to asp-server in some Windows systems.

    Solution

    This issue is fixed in the 1.8.0 image. For images 1.5.x, 1.6.x, and 1.7.x, you can follow the description below to proceed:

    You can modify the server config file and use local socket for connection to prevent the AF_UNIX file being deleted by mistake. Use Cloud Assistant Agent to run the following commands for troubleshooting:

    $server_path = Split-Path -Path (Get-Process -Name aspsvdi).path -Parent
$json_path = $server_path + "\asp-server.json"
$oldLine = '"conn_type": "3",'
$newLine = '"conn_type": "0",'
$content = Get-Content -Path $json_path
$newContent = $content -replace [regex]::Escape($oldLine), $newLine
$newContent | Set-Content -Path $json_path
echo $newContent
taskkill /f /im aspsvdi.exe

    What should I do if end users receive the "Not allowed" message when they access cloud computers deployed in the China (Hong Kong) region and overseas regions from the web client?

    Cause

    When end users use the web client to access cloud computers, the default domain wuying.aliyun.com does not support the access to resources deployed in the China (Hong Kong) region or overseas regions. To resume access, you must bind a custom domain name to the web client in the EDS Enterprise console.

    Solution

    After you bind a custom domain name to the web client, reconnect to the cloud computers. For more information, see Bind a custom domain to a web client.

    Application and peripheral

    What should I do when end users are required to enter an admin password to install applications on their cloud computers?

    If a username and password prompt appears when an end user installs applications, it means the user account does not have local admin privileges on the cloud computer. You can resolve this issue by granting local admin privileges to the end user using one of the following methods:

    • Grant through the Users & Organizations module

      1. Log on to the Elastic Desktop Service Enterprise console.

      2. In the left-side navigation pane, choose Users > Users.

      3. On the User tab of the Users & Organizations page, find the username of the account and choose one of the following methods to grant local admin privileges:

        • Click the 修改.png icon in the Local Administrator column that corresponds to the end user. In the Configure Local Administrator dialog box, select Yes for the Local Administrator parameter and click Confirm.

        • Select the check box of the username in the first column, choose More > Configure Local Administrator, select Yes in the dialog box that appears, and then click Confirm.

      4. Restart the cloud computer.

    • Grant through the Cloud Computers module

      1. In the left-side navigation pane, choose Resources > Cloud Computers.

      2. In the upper-left corner of the top navigation bar, select a region.

      3. On the Cloud Computers page, find the target cloud computer, click the ⋮ icon in the Actions column, and then select Send Remote Commands.

      4. Choose the appropriate command based on the cloud computer OS:

        • Windows

          Set the Command Type parameter to PowerShell and run the following command:

          net localgroup administrators "Username" /add

        • Linux

          Set the Command Type parameter to Shell and run the following command:

          #!/bin/bash 
echo "ecd\\\Username ALL=(ALL) NOPASSWD:ALL #SET_BY_EDS" >> /etc/sudoers
          Note

          • The username should be provided by the end user.

            To find the username, the end user can follow these steps: Connect to the cloud computer. Open Command Prompt (press Win+R, type cmd, and then press Enter). Then, run the whoami command. The returned string is the username.

          • If the end user still encounters a password prompt when installing applications on a Linux cloud computer, proceed as follows: Run the sudo su command to switch to the root user. Run the passwd root command to change the password. Then, retry the application installation.

      5. Restart the cloud computer.

    How do I set self-uploaded applications for automatic installation?

    Whether self-uploaded applications support automatic installation depends on silent installation compatibility.

    • If silent installation is supported, perform the following operations: Locate the target self-uploaded application in the EDS Enterprise console. Click Automatically Install in the Actions column. Then, follow the on-screen instructions to complete the process. For more information, see Manage self-managed applications.

    • If silent installation is not supported, submit a ticket for technical assistance.

    What should I do if the cloud computer fails to recognize the USB device connected to the local terminal?

    If end users cannot access locally connected USB devices on their cloud computers, follow these steps to diagnose and configure the necessary settings:

    Warning

    For Windows cloud computers, avoid installing drivers in the temp directory. If USB drivers are installed in the temp directory, they will be cleared upon reboot, causing USB devices to malfunction.

    1. Check whether USB redirection is enabled. For more information, see Policies for peripherals.

    2. If end users need to use local USB printers or cameras, check whether printer redirection or webcam redirection is enabled accordingly. For more information, see Security-related rules.

    3. If end users need to use a USB security token (U-shield), check whether the necessary drivers for the device are installed.

    4. Ask end users to check whether the USB policy is enabled in their cloud computers.

      Method: On the cloud computer card in the Alibaba Cloud Workspace client, click the 个人中心1.PNG icon in the top-right corner, and choose Settings > Peripherals. Then, configure the USB device whitelist or blacklist as needed.

    Cloud computer startup, restart, and wakeup errors

    What do I do if a Windows cloud computer is stuck "Starting" and then fails, showing as "Stopped"?

    If you encounter this issue, there are several possible causes. Please use the following guide to identify the cause and apply the appropriate solution.

    • Startup failure due to an underlying resource issue

      Cause:

      You are unable to send remote commands to the cloud computer. This may indicate that the cloud computer is in a Blue Screen of Death (BSOD) or frozen state, or that underlying resources failed to allocate due to inventory or virtualization issues.

      Solution:

      Please ask your IT administrator to submit a ticket Alibaba Cloud technical support will resolve the startup failure.

    • Startup failure due to VPN software

      Cause:

      By sending network-related remote commands like ipconfig /all(ifconfig) and route print(route -n), you find network adapters other than Red Hat VirtIO or routes not related to the Ethernet adapter. This may be caused by VPN software installed on the cloud computer.

      Solution:

      Use the taskkill command to terminate the VPN process, or uninstall the VPN software via remote command.

      Note

      If the methods above do not resolve the issue, you can submit a ticket for assistance from Alibaba Cloud technical support.

    • Startup failure due to security software

      Cause:

      By sending the tasklist (ps -aux) remote command to get the list of running processes, you find processes related to security software.

      Solution:

      Use the taskkill command to terminate the security software process, or uninstall the software via remote command.

      If the security software has a management console, add the blocked EDS processes to the whitelist.

    • Note

      If the methods above do not resolve the issue, you can submit a ticket for assistance from Alibaba Cloud technical support.

    • Startup failure due to insufficient disk space

      Cause:

      By sending the remote command Get-PSDrive C | Select-Object Free（df）, you find that the C: drive (system disk) has no free space.

      Solution:

      1. Expand the system disk of the cloud computer in the EDS console.

      2. Delete unnecessary files from the C: drive (system disk) by sending remote commands.

    What do I do if a Linux cloud computer is stuck "Starting" and then fails, showing as "Stopped"?

    Cause: Send the following remote command script to detect and fix the issue.

    #!/bin/bash
echo "1. Checking hostname"
NAME=$(cat /etc/hostname)
echo Local hostname: $NAME"
HOSTNAME=$(curl http://100.100.100.200/latest/meta-data/hostname 2>/dev/null)
echo "Alibaba Cloud Workspace hostname: $HOSTNAME"
if [ "$NAME" != "$HOSTNAME" ]; then
sudo hostnamectl set-hostname $HOSTNAME
echo "Hostname has been changed to $HOSTNAME"
else
echo "Hostnames are identical."
fi
echo "2.Checking resolv.conf"
Raesolv_home=$(cat /etc/resolv.conf | grep "nameserver" | awk '{print $2}')
echo "Raesolv points to: $Raesolv_home"
realm=$(cat /etc/krb5.conf | grep "default_realm" | awk '{print $3}')
if [[ -z $(cat /etc/hosts | grep $realm) ]]; then
HOSTNAME=$(hostname)
CTRL_REALM=$(cat /etc/krb5.conf | grep "default_realm" | awk '{print $3}')
cat <<- EOF >/etc/hosts
127.0.0.1 $HOSTNAME.$CTRL_REALM $HOSTNAME localhost
$(ifconfig eth0 | grep inet | grep -v inet6 | awk '{print $2}') $HOSTNAME $HOSTNAME
AME
EOF
fi
echo "3. Checking DNS IP"
IP=$(cat  /run/systemd/network/10-netplan-eth0.network  |  grep  'DNS='  |   awk -F= '{print $2}' )
DNS_IP=$(curl http://100.100.100.200/latest/user-data 2>/dev/null | grep  '^DNS_IP=' | cut -d "=" -f 2 | cut -d \" -f 2  |  sed 's/,/ /g')
echo $IP
echo $DNS_IP
if [[ "$IP" != "$DNS_IP" ]]; then
        DNS_IP=$(echo $DNS_IP | tr ' ' ',')
        a=`cat /usr/sbin/edsnetworkinit.sh | grep -A 15 eth0 | grep -A 5 nameservers | grep addresses | cut -d : -f 2 | cut -d [ -f 2 | cut -d ] -f 1`
        sed -i "s/$a/$DNS_IP/g" /usr/sbin/edsnetworkinit.sh

fi
/usr/sbin/edsnetworkinit.sh
echo "4. Checking domain"
NAME=$(cat  /run/systemd/network/10-netplan-eth0.network  |  grep  'Domains='  |   awk -F= '{print $2}' )
DOMAIN_NAME=$(curl http://100.100.100.200/latest/user-data 2>/dev/null | grep  $NAME|  cut -d "=" -f 2  |  cut -d \" -f 2)
echo "Local Hostname: $NAME"
echo "EDS domain name: $DOMAIN_NAME"
echo "5. Checking permissions"
for dir in /home/WUYING*
do
    if [ -d "$dir" ]; then
        echo "Found directory: $dir"
        current_owner=$(stat -c '%U' "$dir")
        expected_owner=$(basename "$dir"  | sed 's/_/\\/')
        if [ "$current_owner" != "$expected_owner" ]; then
            echo "Changing owner of $dir from $current_owner to $expected_owner"
            chown -R "$expected_owner" "$dir"
        else
            echo "Owner of $dir is already $expected_owner"
        fi
    fi
done
echo "6. Checking SMB service configuration and domain join status"
smb_conf="/etc/samba/smb.conf"
krb5_conf="/etc/krb5.conf"
smb_realm=$(cat /etc/samba/smb.conf | grep "^realm" | awk '{print $3}')
krb_realm=$(cat /etc/krb5.conf | grep "default_realm" | awk '{print $3}')
# Check if smb.conf contains all specified configurations
if grep -q "workgroup * = *WUYING" $smb_conf &&
   grep -q "template shell * = */bin/bash" $smb_conf &&
   grep -q "template homedir * = */home/%D_%U" $smb_conf && 
   [ "$smb_realm" = "$krb_realm" ]; then
    echo "Not modified"
else
    echo "Modified"
    # Find the appropriate domain join script
    sh=$(find /usr/local/share/aliyun-assist/work/script -type f -size +10000c)
    $sh > /dev/null 2>&1
fi
if net ads testjoin | grep -q "Join is OK"; then
    echo "Domain join successful"
else
    echo "Domain join failed"
fi
echo "7. Checking for black screen issue (libasp-server.so file)"
if [ -n "$(find /opt -name "libasp-server.so*" 2>/dev/null)" ]; then
    echo "libasp-server.so file exists"
else
    echo "libasp-server.so file is missing"
fi
echo "8. Checking firewall and port 5912"
if [ "$(systemctl is-active firewalld)" = "active" ]; then
    echo "Firewall is active, stopping it now..."
    systemctl stop firewalld
    systemctl disable firewalld
else
    echo "Firewall is inactive."
fi
if [ "$(ss -tulpn | grep ':5912')" ]; then
    echo "Port 5912 is open."
else
    echo "Port 5912 is not open, opening it now..."
    firewall-cmd --zone=public --add-port=5912/tcp --permanent
    firewall-cmd --reload
    echo "Port 5912 has been opened."
fi

    Solution:

    1. Address any issues found in the output of the diagnostic script.

    2. Restore the cloud computer from a snapshot or rebuild the system disk.

    Note

    If the methods above do not resolve the issue, you can submit a ticket for assistance from Alibaba Cloud technical support.

    What do I do if a cloud computer is stuck "Starting" after wake-up, then shows as "Hibernated"?

    Cause:

    This may be caused by an anomaly in the cloud computer's wake-up interface, leading to a wake-up failure.

    Solution:

    Try waking the cloud computer multiple times.

    Note

    If the cloud desktop consistently fails to wake up, you can submit a ticket for assistance from Alibaba Cloud technical support.

    What do I do if a Windows cloud computer fails to load the desktop and shows "The trust relationship between this workstation and the primary domain failed"?

    Cause:

    This may occur if the cloud computer object has been deleted from the domain controller, or if it failed a security check with the domain controller.

    Solution:

    You can create a custom image from the affected cloud computer and then use this image to provision a new one. Alternatively, you can directly re-provision the cloud computer using the original base image.

    Note

    If the problem persists after changing the image, you can ticket for assistance from Alibaba Cloud technical support.

    Extension and others

    How can I troubleshoot EDS API call errors?

    If an API call fails, you can use the returned request ID to check the error details on the OpenAPI problem diagnosis platform.

    Note

    An API request involves building the request on the client side, sending it over the network, and processing it on the server. Since errors can occur at any stage, it's recommended to add exception handling logic that fits your business needs.