The short answer is no. But if you take a minute to explain what you really want to achieve, there may well be a way.
Our agent configuraitons don't allow for such changes in the GUI. The local scheduler task is either on or off. So if you want something else, you turn it off and customize your own local scheduler task.
This document has a good example:
The reason is that our inventory service is very unstable and locks up typically twice daily - because the process does not crash and close, we have no way of detecting the lock up and restarting the service other than checking the last update of all the clients using the console (i am waiting on ld support for a solution).
This is causing problems with laptops who rarely appear on the network failling to update the db.
If clients were scanning directly to the ldscan folder, then if the service crashed, the scan files would still be inserted, just a bit late.
That would not work because when the inventory service locks up - no scan files are stored. Hence wanting clients to use the /O command.
Incidentally, i already run ldms_core daily - how could i get that to do what you have said - there is nothing documented regarding restarting the inventory service?
(i am waiting on ld support for a solution)
Your best bet and use of your resources is to work with support on why the service is not working correctly. Escalate the issue through your ESP or LANDesk contact.
In the mean time it my be a handle/memory leak or load issue. Maybe setup a scheduled task to restart the service daily - a couple of hours before the hangs (if the hangs are random then just pick a time to restart the service). This may help. I have to restart my service due to a leak that we are working on.
(Sorry - just got back from holidays)
If I remember my facts right, it's not a handle/memory leak that Mike has - it's something else. That's particularly aggravating, as "something else" is well said, but usually a royal pain in the behind to find just what's going on .
I'm still working off the usual backlog of things when coming back, but I'm pretty sure that I'll get back to looking at this particular issue and see what we can do here to try and figure out what's causing it.
A Local Scheduler script may do the trick.
The only way to change the command-line (of the shortcut, if that's what you're talking about) would be editing the agent config's MSI via a tool like ORCA (which is free, if hard to find) ... though it'd be easier to just add a (new ?) locally scheduled task with results like this.
Potentially something a bit more sophisticated like the following may help along:
1 - Run inventory scan with the "/O=file" parameter, but output locally somewhere (ideally with a unique name - so "DEVICENAME_DATETIME.scn" or a convention like that may make sense).
2 - Copy/Move the file off to a central repository (this may or may not be LDSCAN - there's benefits to both). If you've got the space somewhere, I'd prefer pointing to a separate server, just to have less reliance on a single point of I/O.
3 - Copy/move the stored files off to the Core in regular intervals.
The problem here would be that the default command-line still needs to talk to the Inventory Service (you'd need to remove the references for "/NTT=Core:5007 /S=Core" from the default command line for it to work (and obviously include "/O=...") - that way the clients don't look whether there's an Inventory Service running on your Core.
LANDesk EMEA Technical Lead
Ah, this is a new and improved ldinv32.exe lockup -- the last one I had to deal with would store scans, it just wouldn't process them.
From the manual: "The alert queue, the scheduler queue, the inventory queue, and the extended device discovery queue are all checked for excessive load. If there are greater than 200 files in any of them, an alert message will be recorded in the event viewer and sent via email. If the inventory queue is backed up, the LANDesk Inventory Service will be restarted in case it has hung."