Configure system parameters

Restrictions and guidelines

With RBAC enabled, only security administrators can configure system parameters.

Procedure

1. On the top navigation bar, click System.

2. From the left navigation pane, select Parameters.

3. Click the System Parameters tab.

4. Configure basic system parameters, customization parameters, customized menu parameters, VNC proxy server parameters, HA parameters, image cleanup parameters, CPU and memory parameters, upgrade parameters, and performance monitoring parameters, as described in "Parameters."

5. Click Save.

Parameters

Basic System Parameters

• Idle Timeout: Specify the idle timeout for the operators. An operator automatically logs out of the system when the idle timeout expires.

• Lockout Period: Specify the time period during which the IP address that an operator uses to access the system is forbidden from logging in to the system. The timer starts for an operator's IP address after the operator fails to log in to the system consecutively for the specified times.

• Max Login Attempts: Specify the maximum number of allowable consecutive login failures. If an operator fails to log in to the system for the specified times, the operator's IP address cannot log in to the system within the specified time period.

• Max Online Operators: Specify the maximum number of online operators. 0 indicates that the number of online operators is not limited.

• Log Level: Select a log level. Options include Error, Warning, and Tip.

• Alarm Lifetime: Specify the lifetime period of alarms. If the alarm lifetime is longer than 7 days, make sure the root partition on the single management host or the mysql-share partition on the stateful failover system has enough space. If you fail to do so, system faults might occur.

• Recycled VM Lifetime: Specify the lifetime period of VMs in the recycle bin. The system automatically deletes a VM in the recycle bin after the lifetime period of the VM expires. The value of 0 indicates that VMs in the recycle bin will not be automatically deleted.

• Temporary File Directory: Enter a temporary directory for uploaded and downloaded files.

• Migration Pause Period: Set the VM pause period for migrating the remaining dirty pages before online migration is completed. As a best practice, do not set this parameter.

• Operation Log Lifetime: Specify the time period within which operation logs cannot be deleted.

• Encoding Mode: Select the encoding mode that the hosts use to report alarms to the alarm server. Options include BASE64, GBK, and UTF-8.

• Intel RDT Noisy Neighbor Quieting: Select whether to enable Intel RDT noisy neighbor quieting. After you enable this feature on the system parameter configuration page, you can enable this feature for a VM on the VM editing page. To enable Intel RDT noisy neighbor quieting for a VM, make sure the physical CPUs on the host where the VM resides support this feature.

• Max Snapshots: Specify the maximum number of external snapshots that can be created for a VM. The default value is 2, and the maximum value is 32. When you create snapshots through a snapshot policy, the allowed number of snapshots depends on the smaller value between the Max Snapshots in the snapshot policy and Max Snapshots system parameter. When CAS automatically executes a snapshot policy, if the number of snapshots reaches the limit, it automatically deletes the earliest snapshots. When you manually create external snapshots, the Max Snapshots system parameter takes effect. If the number of snapshots has reached the value for the Max Snapshots system parameter, the system reports an error and prohibits snapshot creation. To manually create a snapshot, delete unnecessary snapshots and try again.

• Max Client Connection: Specify the maximum number of connections that an IP address can establish to CVM.

• Storage Pool Level-2 Directory: Enable the system to create level-2 directories for VMs in storage pools.

  • With this feature enabled, the system automatically creates a level-2 directory named VM name_dir in the root directory of a storage pool to store a VM image file when you perform an operation that creates the VM image file, such as deploying a VM.

  • With this feature disabled, the system does not create level-2 directories in storage pools, and existing level-2 directories are still available for use.

• Shared Storage Fault Action: Select the action to take when the shared storage of a host fails. Options include Restart Host and Do Not Restart Host. The default is Do Not Restart Host.

• VM IPv6 Address Management: Select whether to enable IPv6 configuration for VMs. If you select Yes, you can configure IPv6 addresses for VMs when you add, Edit, or clone VMs, or deploy VMs through templates. When you view information about a VM, its IPv6 address is displayed.

• DRS VM Startup: Set the state of DRS VM startup. With this function enabled, the system checks whether a host meets the VM startup requirements before starting up a VM on the host. If the requirements are not met, the system migrates the VM to the optimal host to start up. This function takes effect only when automatic migration is enabled for VMs.

• Shared Storage Overcommitment Limit: Select whether to enable shared storage overcommitment limit. With this feature disabled, shared storage overcommitment is not limited. As a best practice, enable this feature for critical services. With this feature enabled, you can set the shared storage overcommitment ratio. If you do not enable this feature, monitor the shared storage pool usage. When the shared storage pool usage is high, scale up the pool or delete unnecessary files to avoid read/write service interruption.

• Shared Storage Overcommitment Ratio: Set the shared storage overcommitment ratio. The overcommitment ratio is essentially a multiplier. Setting it to 0 indicates that overcommitment is not allowed. The higher the ratio value, the higher the allowed overcommitment multiplier. After you set the overcommitment ratio, if shared storage overcommitment alarms are not enabled, the system will automatically enable this alarm. The calculation of the overcommitment ratio is based on the data from the management platform, not the value obtained from the df -h command at the CLI.

• Overcommitment Ratio for Shared File Systems: Set the overcommitment ratio for shared file systems. To disable overcommitment, set the value to 0. A larger overcommitment ratio indicates a higher overcommitment multiplier. After you set the overcommitment ratio for shared file systems, the system automatically enables shared file system overcommitment alarm if this feature is not enabled. This parameter is available only after you enable shared file system overcommitment limit.

Customization Parameters

• System Name: Enter a name for the CAS system.

• Copyright Info: Enter the copyright information on the login page.

• Homepage Logo: Enter the path where the homepage logo is saved. The logo is displayed on the top left of the homepage.

• Login Page Logo: Enter the path where the login page logo is saved.

• Homepage Tiny Logo: Enter the path where the small size homepage logo is saved. The logo is displayed on the top left of the homepage when the navigation tree is collapsed.

• Favicon: Enter the path where the favicon is saved.

• Copyright Logo: Enter the path where the copyright logo is saved.

• Login Page Background: Enter the path where the login page background is saved.

Customized Menu Parameters

• Menu Name: Enter the name of the system to be accessed from CVM.

• Menu URL: Enter the URL for accessing the system.

VNC Proxy Server Parameters

• Server IP: Enter the IP address of the VNC proxy server.

• Username: Enter the username used to access the VNC proxy server.

• Password: Enter the password for the user to access the VNC proxy server.

• VNC Console Password: Specify the password used to connect to the VNC consoles of VMs. This password takes effect only on VMs that are created after the password configuration. To change the VNC console password of a specific VM, access the VM editing page.

• VNC Password: Select whether to use VNC random password for the VMs. After you enable this feature, the configured VNC passwords become invalid.

DNS Server Parameters

• DNS Server: Enter a maximum of three semicolon-separated DNS server IP addresses for CVM. This setting takes effect on all hosts on CVM.

HA Parameters

• Storage Heartbeat Interval (Seconds): Specify the interval at which CVK hosts send heartbeat packets to CVM. If a CVK host fails, the heartbeat is interrupted, and VMs on the host are migrated to other hosts in the same cluster after three host heartbeat intervals and three storage heartbeat intervals.

• Storage Heartbeat Interval (Seconds): Specify the interval at which CVK writes timestamps to the storage pool. If a storage network fails, the heartbeat is interrupted, and affected VMs are migrated to other hosts in the same cluster after three storage heartbeat intervals.

• VM Migration Mode: Select a VM migration mode in the HA system. Options include Balance First and Efficiency First.

  • Balance First—Assigns VMs to less-loaded hosts for balanced resource usage. This prioritizes load balancing across nodes in a cluster but might reduce migration speed.

  • Efficiency First—Assigns VMs to more-loaded hosts with shorter migration paths. This prioritizes migration speed and efficiency but might result in uneven load distribution.

Image Cleanup Parameters

• Max Usage (%): Specify the threshold for triggering image cleanup.

  When the space usage of /vms/nova exceeds the maximum usage threshold, the system cleans up the cached images in this path until the space is larger than the reserved space or smaller than the specified space usage after cleanup.

• Reserved Space: Specify the space that must be reserved after image cleanup.

• Usage After Cleanup: Specify the maximum space usage after image cleanup. The value range for this field is 40% to the value for the Max Usage (%) field minus 1%. The default is 60%.

CPU and Memory Parameters

• Reserve Host Memory: Configure whether to reserve host memory. If you select Yes, the system reserves 20% of the host memory or 10 GB host memory (whichever is smaller). The reserved memory of a host cannot be used for VMs started on or migrated to the VM.

• Memory Allocation Ratio for All VMs: Configure the maximum ratio of the memory allocated to all VMs to the total memory size of a host. The system identifies whether this ratio is reached when one of the following events occurs:

  • A VM is created, deployed, or cloned.

  • A VM is migrated manually or automatically by the HA or DRS feature.

  • A VM is restored from the recycle bin.

  • A VM is added by DRX.

  • A VM is migrated to a host through cloud rainbow.

  • CPU or memory allocation is edited for a VM.

  • An OVF template is imported.

  • A VM is restored from a snapshot or backup.

  To ensure service continuity when the maximum memory allocation ratio is exceeded, DRX and VM migration by the HA or DRS feature will still run normally.

• Memory Allocation Ratio for Running VMs: Configure the maximum ratio of the memory allocated to running VMs to the total memory size of a host. The system identifies whether this ratio is reached when one of the following events occurs:

  • A VM starts.

  • CPU or memory allocation is edited for a running VM.

  • A VM is restored from a memory snapshot.

  • A running VM is migrated manually.

  • A VM is migrated automatically by DRS.

  • A VM is added by DRX.

  • A VM is migrated to a host through cloud rainbow.

  To ensure service continuity when the maximum memory allocation ratio is exceeded, DRX and VM migration by the HA or DRS feature will still run normally.

• CPU Overcommit: Select whether to enable CPU overcommitment. Options include Yes and No. If you select Yes, the number of vCPUs bound to the physical CPUs of a NUMA node can exceed the number of the physical CPUs. If you select No, the number of vCPUs bound to the physical CPUs of a NUMA node cannot exceed the number of the physical CPUs.

• CPU Allocation Ratio for All VMs: Configure the maximum ratio of the CPUs allocated to all VMs to the total number of CPUs on a host. The system identifies whether this ratio is reached when one of the following events occurs:

  • A VM is created, deployed, or cloned.

  • A VM is migrated manually or automatically by the HA or DRS feature.

  • A VM is restored from the recycle bin.

  • A VM is added by DRX.

  • A VM is migrated to a host through cloud rainbow.

  • CPU or memory allocation is edited for a VM.

  • An OVF template is imported.

  • A VM is restored from a snapshot or backup.

  To ensure service continuity when the maximum CPU allocation ratio is exceeded, DRX and VM migration by the HA or DRS feature will still run normally.

• CPU Allocation Ratio for Running VMs: Configure the maximum ratio of the CPUs allocated to running VMs to the total number of CPUs on a host. The system identifies whether this ratio is reached when one of the following events occurs:

  • A VM starts.

  • CPU or memory allocation is edited for a running VM.

  • A VM is restored from a memory snapshot.

  • A running VM is migrated manually.

  • A VM is migrated automatically by DRS.

  • A VM is added by DRX.

  • A VM is migrated to a host through cloud rainbow.

  To ensure service continuity when the maximum CPU allocation ratio is exceeded, DRX and VM migration by the HA or DRS feature will still run normally.

Upgrade Parameters

• Graceful VM Shutdown Period (Seconds): Configure this parameter for the system to allow VMs to shut down before host upgrade. This practice can avoid data loss or service interruption on VMs. This parameter takes effect only when the VM task in the upgrade task is set to offline upgrade and the value for VM shutdown is set to auto. By default, the graceful VM shutdown period is 120 seconds.

• Graceful Host Recovery Period (Seconds): Configure this parameter for the system to wait for a specific period of time to all a host to restart successfully and restore to normal state. This practice ensures that the host has completed the upgrade and is running stably before continuing to run VMs. This parameter takes effect only when a restart is required in host upgrade. By default, the graceful host recovery period is 20 seconds.

Performance Monitor Parameters

• Monitor Refresh Interval: Set the refresh interval for real-time performance data. The minimum interval is 30 seconds and the maximum interval is 10 minutes.

• Monitor Data Polling Interval: Set the interval at which the system records performance data. The minimum interval is 10 seconds and the maximum interval is 120 seconds.

• Save To: Specify the directory to which the system saves performance data. This parameter is editable.

• Save Original Data: After you specify a new directory for performance data storage, you can select whether to save the history performance data at the new directory.

• Reset: To restore the performance monitor settings to the default, click Reset. The setting for the Save To field will not be reset.