Startup Error running Correlated Workflows

Replies to This Discussion

Permalink Reply by Marc P. on September 15, 2017 at 8:41am

attaching WorkflowEndpoint log file (here's the exceptions from the log)

2017-09-15 11:08:51.156-04:00 [24] DEBUG - <log4net.Error>Exception during StringFormat: Index (zero based) must be greater than or equal to zero and less than the size of the argument list. <format>  ProcessMessage - Type is Correlated Message Workflow. The ResumeCorrelatedWorkflowCommand command object is {0}. Correlation ID = {1}, Workflow Instance ID = {2}</format><args>{NULL, 63E64D50D14264612C0F6D0663365D42318C0C0E}</args></log4net.Error>

...

2017-09-15 11:08:51.194-04:00 [OrdersEndpoint Runner] DEBUG - 1 commands were queued to be executed.
2017-09-15 11:08:51.194-04:00 [OrdersEndpoint Runner] DEBUG - Dequeing the next command to execute.
2017-09-15 11:08:51.194-04:00 [OrdersEndpoint Runner] DEBUG - Command 1 was dequeued. Executing the command.
2017-09-15 11:08:51.198-04:00 [OrdersEndpoint Runner] DEBUG - The semaphore was released. New value is 99.
2017-09-15 11:08:51.199-04:00 [OrdersEndpoint Runner] ERROR - An exception was reported on the workflow runner thread for workflow endpoint "OrdersEndpoint"
System.NullReferenceException: Object reference not set to an instance of an object.
at NeuronEsb.Server.Runner.Workflow.ExecuteCorrelatedWorkflowCommand.Execute(WorkflowCommandContext context)
at NeuronEsb.Server.Runner.Workflow.WorkflowHostService.RunWorkflows()

Attachments:

• A2IT201106_Enterprise_Workflow Endpoint_OrdersEndpoint-2017-09-15.log, 149 KB

• ▶ Reply

Permalink Reply by Martin Wasznicky on September 15, 2017 at 8:55am

what build are you runnig?

• ▶ Reply

Permalink Reply by Marc P. on September 15, 2017 at 11:09am

3.5.4.1119

• ▶ Reply

Permalink Reply by Scott on September 15, 2017 at 11:49am

This has been fixed in a more recent build so you will want to update to the latest release and also update the database after installing.

• ▶ Reply

Permalink Reply by Marc P. on September 15, 2017 at 2:48pm

thanks Scott. I updated to 3.5.4.1130, the Sample looks to be working ok, but I'm still running into the issue for my solution's Correlated workflow, still getting same error.

But I noticed though I can no longer view a Workflow's activities details from the Workflow Tracking, though it's enabled, and am also seeing odd behavior when trying to restart a workflow, restarting the endpoint (or Neuron service), doesn't appear to be restarting a previous failed workflow it removed, - the workflow on new messages (or existing - they're all just queuing up in pending).

Appreciate any suggestions.

thanks.
Marc

• ▶ Reply

Permalink Reply by Scott on September 15, 2017 at 2:50pm

Can you zip up and then upload your solution?  We don't need the history folder in there so you can omit that.  

• ▶ Reply

Permalink Reply by Marc P. on September 15, 2017 at 3:34pm

sure thing. here's my test solution I've been working with. Also, after reboot I'm now able to see the Workflow activities window again, but still having issue restarting the workflow once it errors out.

thanks,

Marc

Attachments:

• WorkflowTesting.zip, 820 KB

• ▶ Reply

Permalink Reply by Scott on September 15, 2017 at 4:25pm

In your initial receive message activity, put "message" in the activity's "result" property.  When hitting the audit step, the workflow variable "message" is null because it was never set in the receive activity (which is set using the "result" property).

• ▶ Reply