mibuso.com

We are using the Navision job queue for scheduling jobs. Some of the jobs are data ports that will do a straight forward data dumps to a file. The queue runs correctly for most parts of the day, but then during the night some jobs will run much longer than they should or they will process to a point and then stand still. When you long onto the console where the job queue dispatcher is running then all of a sudden the job will start to run again and finish. We now have to logon to the server remotely in the evening to ensure evening jobs don’t hang up.

We have been monitory SQL and there is nothing out of the ordinary that shows up. When the jobs hang up there can sometimes be a few users working on the system or there could be nothing. So it does not appear that the system is overloading with processing.

Our NAVISION server runs on a virtual machine. The environment is as follows:
• VM WARE (V4.1)
• Windows 2008 R2 X64 SP1
• Navision 5.00
• SQL 2008 R2 X64 SP1

Does anybody have an idea why this is occurring or how we can correct this? These so called dead spot do not show up even under windows monitoring? Could this be the virtual environment or a problem on windows itself?

Maybe I misunderstand. Are you saying that you use a NAS to launch the Job Queue to then run a dataport?

No, I use the job queue dispatcher. NAS can not run dataports

Did you bring over the dispatcher form 4.00 then?
There were always problems with this, which was one of the reasons for deprecating it in 5.00 and replacing with NAS.

We started out installation of Nav with version 5.00 and using thje dispatcher. Initially there was no problem with it hanging up. So if I understand you correct this hang up we are experiencing is a problem with the dispatcher?

Sorry I should clarify. There are issues running any code in the back ground, such as OnTimmer in forms, if there is any error or maybe a network interruption or a time out etc, then the process will stop and you need to restart it. Because a From is running unattended, often if something happens you wont see it. Maybe its running on terminal server without any visible console.

The beauty of NAS is that each time you call it you start a new instance of the process again, so if the last one crashed, then next one will generally restart. So using NAS is much more stable than running in a hidden form.

From your description I would guess the code is hitting a time out which pops up a warning message and is sitting there waiting for the user to hit OK.

Let me just clarify quickly the jobs that are running on the despatcher are dataports with no forms popping up other than the progress indicator. Can this indicator be turned off to give no feedback?

Also the jobs dont time out. They stop running until you logon to the console. Then they continue to run

As far as NAS goes we can use that since NAS does not support running of dataports

You need to define the word "Console". |I am starting to think that the issue is related to Terminal Server, or what ever environment you are using to run the Scheduler. It looks like it is going to sleep then waking up when you log in.

The "console" as I call it, is the main session when the server has booted up. In other words the session I log onto with the Administrator account. It seems that it goes to sleep. But I dont understand why as it did not do it in the past. How can I prevent this?

The "console" as I call it, is the main session when the server has booted up. In other words the session I log onto with the Administrator account. It seems that it goes to sleep. But I dont understand why as it did not do it in the past. How can I prevent this?