It sounds like this is not a worklfow problem, but instead an issue with how reverse replication is configured.  Do you have a reverse replication agent setup for each publish instance?  Does the node change that triggers the launcher work on one of the publish nodes but not the other?  If so then y...

Hi Deva,Can you explain a little bit more what you are expecting to see in the logs that is not happening?  From your description it sounds like a workflow was launched and the workflow completed, what else should have happened?Thanks,Will

Hi,Yes, add a configuration for "Apache Sling Logging Logger Configuration"Make sure you log at a debug level.I think the two interesting things to log are:com.adobe.granite.workflow.core.WorkflowSessionImplcom.adobe.granite.workflow.core.jobYou can log all of workflow but you'll see a lot of irrele...

Hi -This is unfortunately not implemented in the touch UI inbox and is a gap between it and the classic inbox.  Please contact support and reference GRANITE-4172 (the internal bug #).thanks and appologies,Will

Hi Julio,The "external" queue is used for special workflow steps which are built as external steps.  External steps are typically used in situations where you are contacting an external system and polling for results.  For example the "InDesign Media Extraction Process" step is implemented as an ext...

Hi,On CQ5.6.1 you can configure this in the OSGI console on the "Adobe Granite Workflow Service".  Change the property "Max Inbox Query Size".That being said, I would suggest purging some of your workflows - you'll get better performance when you keep the number of workflows low.  There is a purge t...

Hi -Can you explain in a bit more detail how you're attempting to accomplish this?If you add a single step at the end of the DAM Update Asset workflow, which does the activation of the asset, it should only be executed a single time.Will

Hello,Yes, stale workflows are in a running state.  This is a little confusing, the basic idea is that the workflow has not finished, and the workflow expects that a Sling Job is going to finish executing the workflow, but that job cannot be found in the system.  Your purge configuration looks corre...

Hi,Glad that was the issue.  Yes the next problem you're seeing also seems to be the same issue.  Basically you want to make sure that any dependencies you use in your POM don't get embedded (either the jar or jar exploded .class files) in the jar you build.  You can open the jar you build in winzip...

Hmm...Is there any chance you are embedding a dependency jar inside your bundle?  If you have your own copy of the javax.jcr.Session class file inside your bundle the classloaders could be messed up.  Sorry again just a guess.Will