Oversized ibdata1
Possible Causes
ibdata1 is the system tablespace of InnoDB. It contains:
- Data related to multi-version concurrency control (MVCC): undo logs
- Metadata of InnoDB tables, such as data dictionaries
- Change buffer and double write buffer
The primary reason for the increase in ibdata1 is excessive undo logs. The reasons for excessive undo logs are as follows:
- Long-running transactions not committed block undo logs from being purged.
- A large number of concurrent writes lead to excessive undo logs which cannot be purged fast enough.
To see the number of undo logs that are not purged, you can check the value of History list length in the command output of show engine innodb status.
Solution
- If the ibdata1 file of the primary instance is too large but that of the standby instance is not, perform a primary/standby switchover.
- If the ibdata1 file of the standby instance is too large but that of the primary instance is not, contact customer service to rebuild the standby instance.
- If the ibdata1 files of both the primary instance and standby instance are large, use Data Replication Service (DRS) to migrate data.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot