Concepts

• Monitoring System:
  A monitoring system provides the following features:

  • Monitor the following two types of time-series data:
    • Resource utilizations such as CPU utilization of VM instances and memory utilization of hosts
    • Resource capacities such as available number of IP addresses and total number of running VM instances
  • Event collection: collects events predefined on the Cloud, such as host disconnection and VM HA enabling.
  • Alarm: triggers alarms on time-series data or events.
  • Audit: records all operations and allows queries.
  • Customization: allows you to customize alarms and message templates and use predefined alarm templates and resource groups.
    • The following three types of alarms are supported:
      • Resource alarm: triggers alarms on time-series data. For example, you can configure an alarm for VM instances. If the CPU utilization of a VM instance exceeds 80% by five consecutive minutes, send an alarm message to an email address.
      • Event alarm: triggers alarms on events, also called event subscription. For example, you can configure an alarm for host disconnection. If a host is disconnected, an alarm message is sent to DingTalk.
      • Extended alarm: receives alarm messages from message sources. For example, if a Ceph Enterprise storage pool is downgraded, an alarm message is sent to the system of the Cloud.
    • A message template specifies the text template of a resource alarm message or event alarm message sent to an SNS system.
      • A message template and message recovery template are provided by the system. If you do not create a template, the system uses the predefined templates.
      • You can create multiple message templates and can set only one template as the default template. Messages are formatted by using the default template.
      • You can use ${} in a template to quote variables configured in an alarm or event.
      • You can configure email, DingTalk, WeCom, Lark, SMS, Webhook, and Microsoft Teams as an endpoint in a message template. Messages sent by using email, DingTalk, WeCom, Lark, SMS, Webhook, or Microsoft Teams are sent in the specified format.
    • A message source is used to take over extended alarm messages. If you configure alarms for message sources, extended alarm messages can be sent to various endpoints. This enables centralized management of alarm messages and improves O&M efficiencies. You can configure a message source to take over alarm messages of Ceph Enterprise.
    • An alarm template is a template of alarm rules. If you associate an alarm template with a resource group, an alarm is created to monitor the resources in the group.
    • A resource group consists of resources grouped based on your business needs. If you associate an alarm template with a resource group, the alarm rules specified by the template take effect on all the resources in the group.
• SNS:

  SNS sends alarm messages to the specified endpoints. The supported types of endpoints include the system, emails, DingTalk, WeCom, Lark, Webhook, SMS, Microsoft Teams, and SNMP trap receivers.

  Endpoints:

  • The system provides the system-type endpoint. If you associate an alarm with this endpoint, alarm messages will be displayed below the Recent Message button in the top right corner of the UI.
  • You can also create an endpoint of the email, DingTalk, Lark, WeCom, Webhook, SMS, Microsoft Teams, or SNMP trap receiver type.

Characteristics

Scenarios

The function of monitoring and alarm monitors the core resources and events of the Cloud platform and sets up an alarm receiving mechanism. When core resources are abnormal, the monitoring and alarm will make real-time responses according to the alarm level to help O&M personnel quickly locate and solve the problem.

Global Setting

• Monitoring data is retained locally for 6 months by default, and you can customize the monitoring data retention period in the basic settings as follows:

  On the main menu of ZStack Cloud, choose Settings > Global Setting > Basic. Then, the Basic tab is displayed. You can set Monitoring Data Retention Period. Enter an integer between 1 and 12. Default: 6. Unit: month.

• Monitoring data is retained locally in a size of 50GB by default, and you can customize the monitoring data retention size in the basic settings as follows:

  On the main menu of ZStack Cloud, choose Settings > Global Setting > Basic. Then, the Basic tab is displayed. You can set Monitoring Data Retention Size based on your needs. Default: 50 GB.

• ZStack Cloud supports receiving extended alarm messages. On the main menu, choose Settings > Global Setting > Advanced. Then, the Advanced tab is displayed. You need to turn on the Extended Alarm Notification switch to use the extended alarm function.

Email Endpoint

DingTalk Endpoint

HTTP Application Endpoint

Short Message Service Endpoint

Microsoft Teams Endpoint

Concepts

• Inspection Categories and Items:
  One-Click Inspection provides five inspection categories, including platform, compute resources, network resources, storage resources, and global settings. You can use the service to inspect key resources and services of the Cloud, such as the management node, hosts, VM instances, image storage, primary storage, physical and virtual NICs and networks, and licenses.

  • Platform: Check the basic services and running status of the Cloud.
  • Compute: Check the usage and running status of physical and virtual compute resources of the Cloud.
  • Network: Check the configurations and status of physical and virtual networks of the Cloud.
  • Storage: Check the usage and running status of physical storage resources of the Cloud.
  • Global Setting: Check the configurations of key resources of the Cloud.

  After you select items from certain inspection categories and launch inspection, related resources or services are inspected and their healthiness is scored. For more information about inspection items, see Inspection Items.

• Inspection Results:
  One-Click Inspection provides four inspection results, including Normal, Warning, Fault, Failed.

  • Normal: The inspected resources or services are in normal status. This result is marked with a green icon.
  • Warning: The health status of inspected resources or services is compromised, which may to some extent affect their performance and stability. This result is marked with a yellow icon.
  • Fault: The inspected resources or services are in critical condition and may seriously affect business operations. This result is marked with a red icon.
  • Failed: The inspection on related resources or services fails, which may seriously affect business operations. This result is marked with a grey icon.
• Healthiness Scoring:

  One-Click Inspection provides an in-built healthiness scoring mechanism for Cloud resources and services. It allows you to grasp the overall running status of the Cloud in a visualized way.

  Scoring on inspected resources/services: Scores resources and services based on the inspection results of related resource and service attributes.

  • If all attributes of a resource or service under inspection are in Normal status, the inspection result of the resource or service is Normal. The score is 100 points.
  • If one attribute of a resource or service under inspection is in Warning state and the other attributes are in Normal status, the inspection result of the resource or service is Warning. The score is 50 points.
  • If one attribute of a resource or service under inspection is in Fault or Failed state, the inspection result of the resource or service is Fault or Failed. The score is 0 points.
  Scoring on inspection items: Scores inspection items based on the inspection results of related resources and services.

  • If an inspection item does not belong to the Global Setting category, the inspection item is scored based on the following mechanism:

    • Score of Inspection Item = (Score of Resource 1 + Score of Resource 2 + …… + Score of Resource N)/(N*100)*100
    • For example, if an inspection item involves 3 resources, which are in Normal, Warning, and Fault/Failed status respectively, the scores of the three resources are 100, 50, and 0 points respectively. Then the score of the inspection item is (100 + 50 + 0)/(3*100)*100=50 points.
  • If an inspection item belongs to the Global Setting category, the inspection item is scored based on the following mechanism:

    • The score of the inspection item is the score of the involved global setting.
    • For example, if the inspection result of the involved global setting is Warning, the score of the global setting is 50 points. Then the score of the inspection item is 50 points.
  Scoring on the Cloud: Scores the Cloud based on the scores of all inspection items.

  • Score of the Cloud = (Score of Inspection Item 1 + Score of Inspection Item 2 + …… + Score of Inspection Item N)/(N*100)*100
  • For example, if you select 3 inspection items, which is scored 100 points, 50 points, and 0 points respectively, then the score of the Cloud is (100 + 50 + 0)/(3*100)*100=50 points.
• O&M Suggestions:

  If resources and services are detected in Warning or Fault status, One-Click Inspection analyzes the hidden dangers and their effects on these resources and services, and provides suggestions on O&M. For more information, see Inspection Items.

• Inspection Reports:

  One-Click Inspection allows you to export PDF-formatted inspection reports. An inspection report summarizes platform configurations, resource status, and inspection results. It also provides details of all abnormal inspection items and corresponding O&M suggestions.

Benefits

Typical Backup Scenarios

Backup Service can be applied to the following three typical scenarios: local backup, remote backup, and Public Cloud backup.

Note: If you have the Tenant Management Plus license at the same time, the project members (project managers, project admins, and general project members) can perform local backup for VM instances, elastic baremetal instances, and data volumes in the project.

• Local Backup

  A local ImageStore image storage can act as the Local Backup Server to store scheduled backup data of the local VM instances, elastic baremetal instances, data volumes, and management node databases. Meanwhile, the seamless switchover between the primary local backup server and the secondary local backup server is supported, which effectively ensures your business continuity.

  If your local data is mistakenly deleted, or data in the local primary storage is damaged, you can recover the backup data from the local backup server, as shown in Local Backup Scenario-1

  
  
  

  If you encounter a disaster in your local data center, you can rely totally on your local backup server to rebuild your data center and recover your business, as shown in Local Backup Scenario-2.

  
  
  

• Remote Backup

  A storage server in a remote data center can act as the Remote Backup Server to store the scheduled backup data of the local VM instances, elastic baremetal instances, volumes, and databases. The backup data needs to be synchronized to the remote backup server from the local backup server.

  If your local data is mistakenly deleted, or data in the local primary storage is damaged, you can recover the backup data from the remote backup server, as shown in Remote Backup Scenario-1

  
  
  

  If you encounter a disaster in your data center, you can rely totally on your remote backup server to rebuild your data center and recover your business, as shown in Remote Backup Scenario-2.

  
  
  

• Public Cloud Backup
  The storage server in the Public Cloud can act as the Public Cloud Backup Server to store the scheduled backup data of the local VM instances, volumes, and databases. The backup data can be synchronized to the Public Cloud backup server from the local backup server.

  Note: The Public Cloud backup feature does not apply to elastic baremetal instances.

  If your local data is mistakenly deleted, or data in the local primary storage is damaged, you can recover the backup data from the Public Cloud backup server, as shown in Public Cloud Backup Scenario-1.

  
  
  

  If you encounter a disaster in your data center, you can rely totally on your Public Cloud backup server to rebuild your data center and recover your business, as shown in Public Cloud Backup Scenario-2.

  
  
  

• ZStack Cloud allows you to integrate 3rd-party backup services via CBT interfaces. For details, contact the official technical support.

Concepts

• CDP task: You can create a CDP task to continuously back up your VM data to a specified backup server to achieve continuous data protection and recovery.
• Recovery task: A recovery task helps you quickly restore data by specifying a CDP task and recovery point, and allows you to view the recovery progress and logs in a more friendly way.
• CDP data: The backup data generated from continuous data protection on VM instances is stored in local backup servers.
• Recovery point: A recovery point is a data point generated during continuous data protection. A recovery point corresponds to a data record within the recovery point interval specified by the user.
• Locked recovery point: You can lock or unlock a recovery point as needed. After a recovery point is locked, data of the recovery point will not be automatically cleared or deleted.

How CDP Works

ZStack Cloud provides block-level continuous data protection. You can restore CDP data according to a specified time point.

• CDP backup

  After continuous data protection is performed on a VM instance, the CDP server first performs a full copy of the VM data, continuously captures I/O data changes, and timestamps, and saves the I/O data of each change, thereby achieving continuous data backup.

  
  
  

• CDP recovery

  When data recovery is performed, the CDP server exposes the CDP data as a block device and then restores the I/O data at a specified time to the disk or file system of a primary storage.

  
  
  

Advantages

• Simple:
  • The software-defined solution we provide is hardware-independent and scalable.
  • You can preview the backup files without restoring the system. Supported file formats: pictures (.png, .jpg, .bmp, .gif, etc.), PDF files, and text files (equal to or smaller than 10 MB).
  • When you create a CDP task for the first time, the Cloud intelligently recommends the desired capacity required by a CDP task based on an algorithm, helping you to plan the backup space reasonably.
  • You can restore data through a wizard-style process.
• Strong:
  • Agentless backup: To use the CDP service, you do not need to install agents for your VM instances or couple with other applications. This helps reduce the configuration complexity, lower the VM performance loss, and guarantee the business security.
  • Second-level RPO: Provides second-level fine-grained continuous data protection for VM instances.
  • Instant recovery: Supports instant recovery with the RTO in seconds, which helps to ensure the business continuity.
  • Primary storage support: The CDP service applies to VM instances in different primary storage scenarios, including local storage, NFS, SharedBlock, Ceph, and CBD.
• Flexible:
  • Flexible RPO settings: Provides second/minute-level RPO settings.
  • Multiple recovery levels: Supports entire recovery and file-level recovery.
    • Entire recovery: You can restore data to the original VM instance or to a newly-created VM instance.
    • File-level recovery: You can retrieve files without restoring the system. Both Windows and Linux file system formats are supported.
  • Flexible data display and search:
    • The CDP data page displays hourly data changes of a VM instance, which provides a reference for backup capacity planning.
    • The CDP data page also provides a recovery point calendar, which identifies the dates with recovery points with colors. This helps you locate a recovery point quickly.
• Reliable:
  • Unified O&M: You can view the critical CDP information on the CDP overview page, including the CDP task status, recovery task status, backup server usage, and unread CDP alarms.
  • You can restore CDP data to the original VM instance by creating a volume for the VM instance. All of the volumes before recovery can be retained and attached to the VM instance again, which ensures the data security to the maximum extent and facilitates post-fault analysis.
  • The "Create VM instance" recovery policy allows you to create a new VM instance from the selected recovery point without affecting the original VM instance. You can finish the recovery after you confirm that the data is correct. This helps to meet the recovery drill requirements.
  • You can mark and lock recovery points to retain the recovery point data in long term.
  • The Cloud provides a list of recovery tasks, allowing you to view the recovery records and progress in a more friendly way.
  • The RPO latency policy and alarm help to effectively relieve data transmission pressure of backup servers in heavy I/O scenarios.

Scenarios

Limits

Currently, if a VM instance has volumes not stored on the Ceph primary storage, you could not clone the VM instance or create snapshots or images for the VM instance during CDP.

Considerations

• Admin tags are created and owned by the administrator while tenant tags are created and owned by tenants.
• Tags created by tenants can only be attached to resources of the corresponding tenants, while admin tags can be attached to all of the resources on the Cloud.
• The administrator can detach or delete tenant tags.
• Tags in a project are owned by the project. Therefore, everyone in the project, including the project admin, project manager, and project member, can perform operations on these tags.
• Currently, tag owners cannot be changed.
• When you change a resource owner, all tenant tags attached to the resource will be detached. However, the admin tags are not affected.
• After the Cloud is upgraded seamlessly, the existing tags will be updated accordingly and displayed in the latest way. If an exception occurs, refresh your browser or create a new tag.

Source Cloud Platform: VMware

Source Cloud Platform: KVM