This may be the wrong place to post about the whole automated patching process using LPM thing (apologies if it is), but I'm not entirely sure where the problem lies, unfortunately. So here's the background:
I have LANDesk 8.8 and Process Manager 4.1, both fully "rolled up" with updates as far as I am aware. I have recently been working with the LANDesk technical gurus to get an automated workflow up and running (patch -> test group, then patch -> pilot group, then patch -> all computers), and have checked the box marked Enable automated patch deployment under the Scan and Repair -> Download Updates dialog box. With me so far?
I have a test group set up (identified via a query) of two computers, one XP Pro VM and one Windows 2000 VM, and I know that they each need a metric shed-load of patches. So I'm waiting with baited breath to see all kinds of exciting action.
But what's actually happening?
I'm getting e-mail notifications on most days (but not all), entitled "Exception notification - patches failed to deploy to the test devices." When I poke around in my LANDesk core, I see a scheduled task called "Patch_Test_<date>_<time>" with my two test group machines listed. Typically, the XP one will show "Unable to get scan and repair settings from core" (My first guess is that this might be a firewall issue, so I turned off the XP firewall today and we'll see what happens there). Meanwhile, the Windows 2000 machine displays "All patches failed." If I scroll along to the right side of the table, the details column shows, "Status = Failed. Details = All tasks completed."
So what am I doing wrong?
I'm only seeing one or two patches that have been set to autofix, so though I believe that the whole LPM workflow deal is being triggered and working, I'm not totally sure. It appears to me that there's something seriously not working with scanning and detecting what's needed on these machines in the first places, and then nothing seems to be effectively getting fixed anyway. Any help at this point would be most appreciated!
Thanks,
Gavin
P.S. Oh hang on, the patches that are set to autofix are the ones that have appeared since I got the automated magical workflow wotnot set up. So it looks like that's working nicely (i.e. they're being found as needed by test group machines, downloaded and set to autofix -- spiffy!). The new question regarding that is, how do I get patches that are older into the mix? Delete them from from my server so they're discovered and downloaded in a new scan? Manually go through the list and set each one to autofix? Any recommendations there? And of course, that still doesn't explain why the actual installation of the new autofix patches to my test group are failing though.