このページは現在英語のみで、表示されている言語を準備しています。
关闭
Tencent Cloud のお知らせ > お知らせの詳細
[TencentDB for MySQL] Announcement on the Storage Space Overuse Locking Rule and Updates to Storage Space Monitoring Metrics
2026-09-04 11:41:33
Dear Tencent Cloud user,
To optimize the monitoring metrics for TencentDB for MySQL storage space (disk), enhance the transparency of storage composition, standardize resource usage, and ensure database service quality, Tencent Cloud will update the storage overuse locking rules for instances, storage space monitoring metrics, and event alarms. The relevant details are announced as follows:
I. Scope of the Update
Item | Description |
Affected Instance Types | TencentDB for MySQL Local Disk Edition: Two-node (General/Dedicated/Economy), Three-node (General/Dedicated), Four-node (General/Dedicated), Read-Only Instance (General/Dedicated), and Disaster Recovery Instance (General/Dedicated). |
Update Details | 1. Storage space overuse locking rules. 2. Storage space monitoring metrics, console display, and event alarms. |
II. Updates to the Storage Space Overuse Locking Rule
2.1 Rule Description
Local disk instances adopt a compute-storage integrated architecture, where the storage space (disk) is bound to the physical host, imposing a hard capacity limit. After this update, when the data file size of a local disk instance exceeds the purchased storage space (i.e., storage overuse), the system will enforce control in two phases: Pre-lock → Lock:
Management Stage | Trigger Condition | Instance Status | Notification Method |
Pre-lock | Data file size exceeds the purchased storage space | Read/Write available | Trigger the "Instance will be pre-locked" event alarm |
Lock | The instance remains in the storage overuse state 24 hours after being pre-locked | Read operations are allowed; write operations will be rejected | Trigger the "Disk Usage Overlimit Lock" event alarm |
Note：
After being locked, the instance only supports read operations; business write operations will fail. It is recommended that you complete storage space expansion or data cleanup within 24 hours of receiving the pre-lock notification to avoid impacting your business. At the same time, please ensure that your application has database connection retry and write failure handling capabilities.
If the instance remains in the storage expansion task execution state after exceeding the pre-lock event time window (24 hours), the system will not enforce read-only mode on the instance. Instead, it will wait until the task is completed before evaluating whether to trigger the lock event.
For instances that were already in a storage overage state prior to the effective date of this announcement, we will contact you through your pre-sales team, key account team, or account manager to arrange specification upgrades or data cleanup. We will also notify you of the affected instance scope and provide operational guidance via email, in-site messages, and SMS. Instances that remain in an overage state after the effective date will be locked. For instances that are not in an overage state, starting from the effective date of this announcement, when data files exceed the purchased storage space, pre-locking and locking will be enforced in accordance with the rules outlined in this announcement.
2.2 Lock Release Event
Scenario | Trigger Condition | Result |
Pre-lock lifted | During the pre-lock period, the data file size returns to within the purchased storage space (e.g., by initiating/completing an expansion or cleaning up data) | Trigger the "Instance Pre-lock Lifted" event alarm; the instance will not enter the lock phase |
Lock resolved | During the lock period, the data file size returns to within the purchased storage space (e.g., by completing an expansion or cleaning up data) | Trigger the "Instance Lock Released" event alarm; the instance resumes read/write operations |
2.3 Additional Notes
If the actual total used storage space of your instance has exceeded the resource usage of the purchased specification, please refer to the Billing Description for Storage Space Overuse for the applicable billing rules.
2.4 Handling Plan for Storage Overuse Lock
After an instance is locked due to storage overuse, it is recommended to unlock it by upgrading the configuration to expand the storage space. If you confirm that there is redundant data that can be cleared, you can also perform a data cleanup operation. For more details, please refer to How to Resolve Storage Overuse Lock Caused by Data Files Occupying All Storage Space.
III. Updates to Storage Space Monitoring Metrics and Event Alarms
3.1 Added Storage Space Metrics
The following storage space metrics have been added for local disk instances:
Metric | Definition |
Capacity | The amount of storage space (disk) used by the user |
CapacityRate | (User's storage space (disk) usage / Purchased storage space) * 100% Purpose: Provided to users as an indicator for determining storage overuse billing |
RealCapacity | Total storage space of data files |
RealCapacityRate | (Instance data file storage space (disk) size / Purchased space) * 100% Purpose: Provided to users as an indicator for determining storage overuse locking |
LogFileCapacity | Total storage space of log files |
LogCapacityBinlog | Binlog file storage space |
LogCapacityRedolog | Redolog file storage space |
LogCapacityUndolog | Undolog file storage space |
LogCapacityRelaylog | Relaylog file storage space |
LogCapacitySlowlog | Slowlog file storage space |
LogCapacityErrorlog | Errorlog file storage space |
LogFileCapacityElse | Storage space of other log files excluding the above |
LogFileCapacityRate | (Total log file storage space / Purchased space) × 100% |
CapacityElse | Instance temporary files and other files |
DiskTmpUsed | Including temporary files generated by temporary tables, sorting operations, etc |
OtherFileCapacity | Other unclassified files |
RemainCapacity | Purchased space - Storage space usage |
RemainCapacityRate | ((Purchased space - Storage space usage) / Purchased space) * 100% |
3.2 Changes and Impact Description of Existing Monitoring Metrics
The impact on existing metrics is as follows:
Disk Utilization (VolumeRate): This metric will be gradually phased out through a gray release. Currently, this metric is calculated as (Data files / Purchased storage size) * 100%, and as ((Data files + Binlog files) / Purchased storage space) * 100% for whitelisted instances.
Disk Remaining (DiskRemaining): This metric is being gradually deprecated (gray release). Currently, this metric is linked to Disk Utilization (also being gradually deprecated) as follows: Disk Remaining = Purchased Storage Space − Data File Space. If you have whitelisted Disk Utilization, then Disk Remaining = Purchased Storage Space − Data File Space − Binlog File Space. Please subscribe to the latest monitoring metric going forward: Remaining Storage Space (RemainCapacity).
Disk Space Usage (Capacity): Only the metric name will be changed to Storage Space Usage.
Data File Usage Space (RealCapacity): Only the metric name will be changed to Data File Usage.
After the new metrics are added, the default monitoring alarms configured for Cloud Database for MySQL will replace the metrics, and the metric values may change to some extent (Disk Utilization will be changed to Storage Space Utilization, and Data File Space Utilization). Please be aware of this:
If you have already configured alarm policies based on legacy metrics such as Disk Utilization and Disk Space Usage, it is recommended that you re-evaluate your alarm metrics and thresholds after this update to avoid false or missed alarms.
Adjustments to metric definitions only affect the display and statistics of monitoring data, and do not impact the actual running status or data of the instance.
The formal deprecation of Disk Utilization (VolumeRate) and Disk Remaining (DiskRemaining) will be communicated to users again via an announcement and/or the console.
3.3 Added Event Alarms
To support the storage space overuse locking rule, the following event alarms have been added in this update. We recommend that you subscribe to them promptly:
Event Alarm Name | Trigger Condition | Recommended Action |
Instance will be pre-locked | Data file size exceeds the purchased storage space | The instance is still readable and writable. Please expand storage or clean up data as soon as possible to avoid entering the locked state after 24 hours |
Disk usage overlimit lock | Remains in the storage overuse state 24 hours after being pre-locked | The instance is read-only. Please expand storage or clean up data immediately to restore read/write operations |
Instance pre-lock lifted | Data file size returns to within the purchased storage space during the pre-lock period | No action required |
Instance lock released | Expansion or data cleanup is completed during the lock period | Monitor the recovery of business write operations |
IV. Effective Date
This update will officially take effect starting at 11:00:00 on October 15, 2026 (Beijing Time). After the effective time, monitoring metrics, console display, and event alerts will be progressively switched over and completed.
V. Contact Us
If you have any questions, please contact us via the Tencent Cloud Ticket System. Thank you for your understanding and support!
