Long running script execution failure

General Discussion
1

We are unable to execute long-running queries which take more than 2 hours (131 Minutes). After 2 hours of execution, the pipeline job failed with the below error. After increasing the tcp keepalive settings to 4 hours, the pipeline job is not failing after 2 hours of execution but it is continuing its execution even after the DB side execution is completed. Then after 4 hours, the job fails with the same error message. We have tried all db side /jdbc/cicd timeout parameters and none of them worked. Did anyone why this behavior is happening for scripts that takes more than 2 hours?

Liquibase version: 4.5.0
Mysql DB version: 8.0.26
Mysql connector: 8.0.27

Error:

Caused by: com.mysql.cj.exceptions.CJCommunicationsException: Communications link failure
The last packet successfully received from the server was 7,904,008 milliseconds ago. The last packet sent successfully to the server was 7,904,008 milliseconds ago.

2

For Oracle I’ve had applications with deployments over 15 hours without any issues. The timeout issue is likely caused by the pipeline.

3

pipeline and runner timeout settings are already checked. Do we have any specific parameter to check in pipeline settings?

Interesting thing we noticed is that , after modifying tcp keep alive to more than 2 hours, the pile job is not finishing even after the db side execution is complete.