For scheduled PolarDB O&M events, you can be notified by text message, phone call, email, or internal message. You can also be notified in the console. Scheduled PolarDB O&M events include database software upgrade events and hardware maintenance and upgrade events. You can view the details of each scheduled event. These details include the event type, task ID, cluster name, and switching time. You can also change the switching time.

Precautions

  • If you have scheduled O&M events and want to view event notifications, you can go to the left-side navigation pane in the console and choose Event Center > Scheduled Events. Scheduled Events
  • In most cases, you are notified of scheduled events in ApsaraDB at least three days before these events are executed. You can be notified in many ways. For example, you can be notified by phone call, email, or internal message. To use this feature, you must log on to Message Center, select ApsaraDB Fault or Maintenance Notifications, and then specify a contact. We recommend that you specify an O&M engineer as the contact. If you do not specify a contact, you cannot receive notifications.
    Figure 1. Message Center settings
    Message Center settings

Procedure

  1. Log on to the PolarDB console.
  2. In the upper-left corner of the console, select the region in which the cluster that you want to manage is deployed.
  3. In the left-side navigation pane, choose Event Center > Scheduled Events.
    Note If a scheduled event requires you to schedule the time to handle the event, a message appears, which prompts you to schedule the time at your earliest opportunity.
  4. Optional:On the Scheduled Events page, you can specify periodic switching time.
    1. In the upper-right corner, click Global Scheduler.
      Periodic switching time
      Note The Global Scheduler panel provides the global configuration items of scheduled O&M events (excluding high-risk vulnerability fix events). After the periodic switching time is set, the scheduled switching time of the new scheduled O&M event will automatically use the periodic switching time. If the periodic switching time is not set, the scheduled switching time of the new scheduled O&M event will automatically use the maintenance window of the cluster.
    2. In the Global Scheduler dialog box, configure the periodic switching time and click OK.
      Global Scheduler
  5. On the Scheduled Events page, you can view the details of the event. To change the switching time of the event, select the cluster that you want to manage and click Add Scheduled Event.
  6. In the Add Scheduled Time dialog box, configure Scheduled Switching Time and click OK.
    Note
    • If you select Earliest Execution Time, the system automatically enters the earliest execution date and time. After you click OK, the events in the cluster are pending for processing. If you clear Set the earliest execution time, you can change the scheduled switching date and time.
    • The time that is specified by the Scheduled Switching Time parameter cannot be later than the time that is specified by the Start Deadline parameter.

Causes and impacts of events

Upgrade type Cause Impact Description
Hot update Cluster migration Transient connections
When the switching is performed at the Scheduled switching time you may experience the following impacts:
  • Typically, you can use the hot upgrade mode to upgrade the minor version of a cluster. Your clusters or data shards in your clusters experience transient connections and stay in the read-only state for up to 30 seconds before all the data is synchronized. We recommend that you perform the switching during off-peak hours. Make sure that your application can be automatically reconnected to your database system.
  • You cannot manage your clusters by using Data Management (DMS) or Data Transmission Service (DTS). This impact is temporary.
Switching between primary and read-only nodes
Cluster parameter adjustment
Host vulnerability fixing
SSL certificate update
Backup mode change
Minor version update Transient connections
When the switching is performed at the Scheduled switching time you may experience the following impacts:
  • Typically, you can use the hot upgrade mode to upgrade the minor version of a cluster. Your clusters or data shards in your clusters experience transient connections and stay in the read-only state for up to 30 seconds before all the data is synchronized. We recommend that you perform the switching during off-peak hours. Make sure that your application can be automatically reconnected to your database system.
  • You cannot manage your clusters by using Data Management (DMS) or Data Transmission Service (DTS). This impact is temporary.
Differences between minor versions

Different minor versions have different updates. You must check the differences between the current minor version and the minor version to which your nodes are updated. For more information, see Release notes.

Minor version update for proxy nodes Transient connections
When the switching is performed at the Scheduled switching time you may experience the following impacts:
  • Typically, you can use the hot upgrade mode to upgrade the minor version of a cluster. Your clusters or data shards in your clusters experience transient connections and stay in the read-only state for up to 30 seconds before all the data is synchronized. We recommend that you perform the switching during off-peak hours. Make sure that your application can be automatically reconnected to your database system.
  • You cannot manage your clusters by using Data Management (DMS) or Data Transmission Service (DTS). This impact is temporary.
Differences between minor versions Different minor versions have different updates. You must check the differences between the current minor version and the minor version to which your nodes are updated.
Network upgrade Transient connections
When the switching is performed at the Scheduled switching time you may experience the following impacts:
  • Typically, you can use the hot upgrade mode to upgrade the minor version of a cluster. Your clusters or data shards in your clusters experience transient connections and stay in the read-only state for up to 30 seconds before all the data is synchronized. We recommend that you perform the switching during off-peak hours. Make sure that your application can be automatically reconnected to your database system.
  • You cannot manage your clusters by using Data Management (DMS) or Data Transmission Service (DTS). This impact is temporary.
Change of virtual IP addresses Some network upgrades may involve cross-zone migrations, which change the virtual IP address of a cluster. If a client uses a virtual IP address to connect to a cloud database, the connection is interrupted.
Note To prevent transient connections, you must use the endpoint in the form of a domain name that is provided by your cluster and disable the DNS cache feature of the application and its server.
Storage gateway upgrade I/O jitter Temporary I/O jitter may occur, and the SQL latency may increase. These impacts last no longer than three seconds.
Cold upgrade Major version upgrade Transient connections
When the switching is performed at the Scheduled switching time you may experience the following impacts:
  • Your clusters or data shards in your clusters experience transient connections and stay in the read-only state for up to two minutes. If your instances contain a large number of table files, are executing large transactions before the upgrade, or encounter high CPU utilization before the upgrade, transient connections usually exceed two minutes. We recommend that you upgrade the version of a cluster during off-peak hours and ensure that your application supports automatic reconnection.
  • If a cold upgrade is required, check the differences between versions and select an appropriate upgrade time based on your business situation.
Note To ensure data security, we recommend that you back up your data before you upgrade major versions or upgrade public preview versions.
Public preview version upgrade

Related API operations

Operation Description
DescribePendingMaintenanceActions Queries the numbers of scheduled events for different types of tasks.
ModifyPendingMaintenanceAction Modifies the task switching time of scheduled events.
DescribePendingMaintenanceAction Queries the details about scheduled events.