conductor server hikari pool deadlock

[ghkdwlgns612]

Details
Conductor version: 3.13.6
Persistence implementation: Mysql
Queue implementation: Mysql
Lock: Zookeeper

We are using the Mysql implementation for all DAO interfaces except IndexDAO. The worker sends a polling request to the Conductor-server every one second. Currently, there are one conductor-server and seven or eight conductor-workers in use.
However, the following error occurs when Workflow and Task are not running. The reason is expected to be due to executionDAOFacade.updateTaskLastPoll(taskType, domain, workerId) inside the Polling method.

Any suggestions other than increasing the number of Conductor-servers? I'm thinking of updating only if there is a Polled Task.

09:05:11.982 [http-nio-8080-exec-8] ERROR com.netflix.conductor.core.dal.ExecutionDAOFacade - Error updating PollData for task: sink-worker in domain: null from worker: demo-ndap-ndap-conductor-worker-6c695bd677-ccfz8
com.netflix.conductor.core.exception.NonTransientException: (conn=20936) Lock wait timeout exceeded; try restarting transaction
	at com.netflix.conductor.mysql.dao.MySQLBaseDAO.getWithRetriedTransactions(MySQLBaseDAO.java:147) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLBaseDAO.withTransaction(MySQLBaseDAO.java:192) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLExecutionDAO.updateLastPollData(MySQLExecutionDAO.java:527) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.core.dal.ExecutionDAOFacade.updateTaskLastPoll(ExecutionDAOFacade.java:585) ~[conductor-core-7.2.0-SNAPSHOT.jar!/:7.2.0-SNAPSHOT]
	at com.netflix.conductor.service.ExecutionService.poll(ExecutionService.java:182) ~[conductor-core-7.2.0-SNAPSHOT.jar!/:7.2.0-SNAPSHOT]
	at com.netflix.conductor.service.TaskServiceImpl.batchPoll(TaskServiceImpl.java:89) ~[conductor-core-7.2.0-SNAPSHOT.jar!/:7.2.0-SNAPSHOT]
	at com.netflix.conductor.service.TaskServiceImpl$$FastClassBySpringCGLIB$$d78c3007.invoke(<generated>) ~[conductor-core-7.2.0-SNAPSHOT.jar!/:7.2.0-SNAPSHOT]
	at org.springframework.cglib.proxy.MethodProxy.invoke(MethodProxy.java:218) ~[spring-core-5.3.22.jar!/:5.3.22]
...
Caused by: com.netflix.conductor.core.exception.NonTransientException: (conn=20936) Lock wait timeout exceeded; try restarting transaction
	at com.netflix.conductor.mysql.util.Query.executeUpdate(Query.java:284) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLExecutionDAO.lambda$insertOrUpdatePollData$55(MySQLExecutionDAO.java:1008) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLBaseDAO.query(MySQLBaseDAO.java:224) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLExecutionDAO.insertOrUpdatePollData(MySQLExecutionDAO.java:1001) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLExecutionDAO.lambda$updateLastPollData$24(MySQLExecutionDAO.java:527) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLBaseDAO.lambda$withTransaction$4(MySQLBaseDAO.java:194) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]
	at com.netflix.conductor.mysql.dao.MySQLBaseDAO.getWithTransaction(MySQLBaseDAO.java:121) ~[conductor-mysql-persistence-3.13.3.jar!/:3.13.3]