This isnt really tied into reporting, but I have a VBScript that you can setup to run via GPO that will run an Agent Health Check on a machine, and if it fails any part of the check will reinstall the agent.
The nice thing about it is that its running on an authority outside of LANDesk that has rights to your machines.
The script is here:
It also includes an email notification so that when an agent fails the health check it will send out an email to whoever is configured to get it. You could probabably use that as a data source once you have the script implemented.
Hope this helps (in a roundabout way).
Thanks for the reply, Chad. Unfortunately this workaround won't work for me. I am quite limited in what I am allowed to do. I can effectively Read only.
Maybe this will help. One of the LANDesk operators over here showed me how he can go to a task that has completed (say a task to deploy patches) and open up "All Devices" under that task. He will get a breakdown of the task results for each device that was targeted. A column called Status can show certain values such as:
Cannot Find Agent
I was thinking that this might be stored in the PatchHistory table however it doesn't appear to be the case. Even if it were held there that wouldn't take into account statuses resulting from tasks that were not patch related.
You might ask "why not just use the feature you just mentioned". Well, that's what we have been using. We want people to be able to extrac this data without having to open the LANDesk console. I am building many different reports for our LD environment via SSRS and I need to find what I need from access to the LD DB tables.
Chad's script is pretty cool. I guess you have to determine what constitutes a bad client? Is the local scheduler service not on a bad client, missing local scheduler tasks, incorrect rights, etc... There are some many pieces that if not functioning/installed could be determined as a bad client. IMO Off, Cannot find Agent may not be bad agents. You could create a LANDesk query looking for all the bits to be in place (services installed and running, LSTs installed, etc...) then perform an agent install.
OK, I think I've got it. While there are, undoubtedly, many flags that could constitute a "bad" agent (non running service, mis-configured agent, permissions, etc) I was hoping to grab some low-hanging fruit on some more common problems.
I have to verify this in our production environment but it looks like the data I was looking for in in LD_TASK, LD_TASK_MACHINE, LD_MESSAGE and PACKAGE tables.
Thanks to all who responded.