Issues with external tasks on Postgres cluster

Camunda 7 Topics Discussion & Questions
1

Hello everyone.

Recently our team started actively using Camunda and we noticed that some of the external tasks were failing to complete with following error:

org.camunda.bpm.engine.BadUserRequestException: External Task 3ca93783-16bf-11ec-9264-0242ac110020 cannot be completed by worker 'bpm'. It is locked by worker 'null'.
	at org.camunda.bpm.engine.impl.cmd.HandleExternalTaskCmd.execute(HandleExternalTaskCmd.java:52) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.cmd.HandleExternalTaskCmd.execute(HandleExternalTaskCmd.java:31) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.interceptor.CommandExecutorImpl.execute(CommandExecutorImpl.java:28) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.interceptor.CommandContextInterceptor.execute(CommandContextInterceptor.java:110) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.spring.SpringTransactionInterceptor$1.doInTransaction(SpringTransactionInterceptor.java:46) ~[camunda-engine-spring-7.12.0.jar!/:7.12.0]
	at org.springframework.transaction.support.TransactionTemplate.execute(TransactionTemplate.java:140) ~[spring-tx-5.2.5.RELEASE.jar!/:5.2.5.RELEASE]
	at org.camunda.bpm.engine.spring.SpringTransactionInterceptor.execute(SpringTransactionInterceptor.java:44) ~[camunda-engine-spring-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.interceptor.ProcessApplicationContextInterceptor.execute(ProcessApplicationContextInterceptor.java:70) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.interceptor.LogInterceptor.execute(LogInterceptor.java:33) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.ExternalTaskServiceImpl.complete(ExternalTaskServiceImpl.java:56) ~[camunda-engine-7.12.0.jar!/:7.12.0]
	at org.camunda.bpm.engine.impl.ExternalTaskServiceImpl.complete(ExternalTaskServiceImpl.java:48) ~[camunda-engine-7.12.0.jar!/:7.12.0]

We assume that the reason behind that is a PostgreSQL cluster that we are using. (with pgpool)
Engine saves a task and its parameters, then tries to load it back and complete it, but data has not been replicated yet to a DB from which data is being read.
It happens every time pool runs out of connections and requests a new one.
Using “serializable” transaction level also results in error in Camunda engine.

Is there any solution for this issue?

Thanks.

1 Like