Neuron ESB User Network

The Service Bus for the Connected Business

Here's the setup:

I have a message that comes into a workflow, correlated based on time.  All messages in a certain timeframe are correlated.  While processing a message, I will never know if it is the last one for that particular correlation set.  The indicator of the completion of particular correlation set is receiving a message in the next correlation set.  Because it is time based, there will never be more than a single active correlation set.

Is there a best practice for house cleaning the old correlation instances?  Should I even care?  Does Neuron have some housecleaning functionality I could use to retire dormant instances after X days?

The best I've come up with so far is to add a branch inside the loop that receives messages looking for a special property that will terminate the loop, and therefore should terminate that instance because the loop is the entire flow.  When the workflow detects it is the first run for a new correlation, it will then publish a message to itself with the previous correlation id and the special property to shut it down.

Tags: correlated, correlation, correlationset, workflow

Views: 167

Reply to This

Replies to This Discussion

hi,

if you have a loop with a correlated receive inside of it, you need to have some logic within the loop that determines when to exit the loop so that you don't have a zombie workflow laying around.  for instance, you could have a timeout activity....if no other like message is received within a period of time it exits. 

I have added a Timeout from the Control Flow section of the toolbox, however I don't seem to be able to correctly specify a timeout parameter.

I'm assuming the requested input is HH:MM:SS, ie: 04:00:00 is a timeout value of 4 hours.  Using this value appears to work (in the Explorer) without issue.  However, when I adjust the parameter to a value more correct for our process (240 hours) something replaces 240:00:00 with 240.00:00:00, and the timeout then creates an exception attempting to parse the value.  Is there a known way I can set the value of the timeout to be multiple days long?

RSS

Neuron ESB Product Support Forums and Communities

Latest Activity

Anupama Nair posted a discussion

Marketo Adapter Invalid Token

Hi,We are using the Marketo adapter to push account updates to Marketo. It works well for some time then starts failing with Invalid Token unless restarted. Is there a configuration that can be done so it can auto refresh the token when required?Thanks!See More
Nov 6, 2023
Sayantini Basak posted a discussion

Maximum payload size(REST API) for requests interfacing to NeuronESB

I am new to Neuron ESB and in our current scenario,We need to process batch transactions comprising of ~1000 records and send them to Neuron ESB for further processing. I would like to understand what is the maximum size of payload that can be transferred using REST interface to Neuron ESB.See More
Jul 22, 2022
Profile IconRobert E Dunie and Sayantini Basak joined Neuron ESB User Network
Apr 28, 2022
Profile IconDayanand, Frederic C, Steffen Greve-Oksfeldt and 1 more joined Neuron ESB User Network
Mar 16, 2022
Profile IconCam Vong and Mitja Luznar joined Neuron ESB User Network
Jan 27, 2022
Profile IconWill Hitzges, Chad Parsons, michael larsen and 4 more joined Neuron ESB User Network
Jun 11, 2021
Anupama Nair posted a discussion

ODBC stored proc polling with temporary tables

We have set up an ODBC adapter to poll a stored proc.We found that if the stored proc has a temporary table defined the rows returned are always 0.Any idea why this would be and what we can do to get around it?See More
Dec 14, 2020
Prasanth Kharade is now a member of Neuron ESB User Network
Dec 30, 2019

Badge

Loading…

© 2024   Created by Neuron Admin.   Powered by

Badges  |  Report an Issue  |  Terms of Service