There is no tidb database used, but this asynchronous task （create_tidb_serverless_task） will cause tens of millions of keys with the prefix (tidb_serverless_cluster_password:*) in redis, causing redis memory overflow.

[wangiii]

Self Checks

• This is only for bug report, if you would like to ask a question, please head to Discussions.
• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template :) and fill in all the required fields.

Dify version

0.10.2

Cloud or Self Hosted

Self Hosted (Source)

Steps to reproduce

Pull v0.10.2 and run it to reproduce

✔️ Expected Behavior

Environments that do not use this database should not perform this task.

❌ Actual Behavior

causing redis memory overflow.

[wangiii]

[dosubot]

The issue you're experiencing with the create_tidb_serverless_task causing Redis memory overflow is due to the task generating a large number of keys with the prefix tidb_serverless_cluster_password:*. This happens because each time a new cluster is created, a unique password is generated and stored in Redis with a key that includes the cluster's display name. If the task is running frequently or in large batches, it will create many such keys, leading to memory issues ^[1].

To address this, you can look into the configuration options that control the execution of this task. The execution of create_tidb_serverless_task is controlled by configuration variables such as dify_config.TIDB_SERVERLESS_NUMBER, which determines the number of idle TiDB serverless instances required before breaking the loop in the task ^[2]. Adjusting these settings or disabling the task in environments where TiDB is not used could help prevent the issue.

To continue talking to Dosu, mention @dosu.

[JohnJyong]

In fact, we do not provide celery beat service, so celery's scheduled tasks will not be triggered. Of course, we will add task control switches in the next version to solve this problem. @wangiii