Timer event throws delegate not found

I have a timer event which will do retries for 5 times, we have issue which occurs intermittently, like after few retries it throws delegate not found. Here is the problem it is able to find the delegate before entering the event or after few retires , why is it not possible to find same delegate again.
Like it is able to find the at 3 rd retry, why it is not able to find the same delegate at 4th time.

@Niall

Thanks in advance

@tiesebarrell any help

@aravindhrs any help.
Thanks

@gowtham_m can you upload your BPMN model? I will check that

@gowtham_m Are you running multiple engines and/or an instance of Camunda Webapp somewhere? It might be the case that one of your engines that doesn’t actually have the classes is picking up the timer job (because for some reason, job execution is not switched off, as it probably should be OR because the process doesn’t contain an async marker before the point where the delegate is invoked) and failing because it cannot execute it. This would explain the “random” failing of the job as well as the “automatic recovery”.

@tiesebarrell multiple engines which contains the same artifacts homogeneous cluster… so each node will contain all the required classes

We have seen this behaviour even with one node

Hi Gowtham,

Is this timer issue resolved . I am getting same issue .
I have tried doing all configurations ,jobexecutor = true , jobexecutordeploymentaware= true .
But none work.

Thanks,
Sonali

@tiesebarrell @Niall @aravindhrs

below: is the bpm-platform.xml config, still i am facing delegate not found, after execution of timer event. It is a homogenious cluster

default org.camunda.bpm.engine.impl.cfg.StandaloneProcessEngineConfiguration java:jdbc/ProcessEngine
<properties>
  <property name="history">full</property>
  <property name="databaseSchemaUpdate">true</property>
  <property name="authorizationEnabled">true</property>
  <property name="jobExecutorDeploymentAware">true</property>
</properties>

Can you post a stacktrace? That might provide a clue as to why it’s not finding the delegate.

As you stated that you’re using Homogeneous cluster setup, then the property jobExecutorDeploymentAware=true is not applicable.

For homogeneous cluster setup, it should be always jobExecutorDeploymentAware=false and it’s default behaviour.

When you set “jobExecutorDeploymentAware=true”, the jobs will be always execute the jobs belongs to the deployed process at specific instance.

Hi @aravindhrs , We are also facing same issue ‘when using time event, getting delegate class not found exception’ [only for few times not all times]. We have heterogeneous deployment [multiple microservices point to same Camunda DB]. We have already set camunda.bpm.job-execution.deployment-aware = true.

Pls advice.

Camunda engine Version : 7.12.0
DB : mariaDB
Deployment : SpringBoot MS
Containter : Openshift

all the services are enabled with camunda.bpm.job-execution.deployment-aware = true with this property?