HCHTech
Well-Known Member
- Reaction score
- 4,138
- Location
- Pittsburgh, PA - USA
We did a server migration for a client recently, and because we wanted to maintain the same drive letters for the mapped drives from the old server to the new, I used the "Replace" option in my mapped drive group policy. So far, so good, right? It worked l like a charm, everyone got the change and all of the 6 different mapped drives kept their same letters but now pointed to the new server instead of the old.
What I should have done, though, was edit this policy so that the mapping was set to "Update" after the migration was finished. Because I missed this step, we chased a problem for over a week with one of their LOB softwares kicking out users if they left the software up for an extended period of time. What was happening was that the "replace" policy was running (the schedule is different for every user since they don't all log into their computers at the same time.) When the policy ran, it momentarily disconnected the drives, then reconnected them, which made the LOB software think the network went down, so it crashed and the user was unceremoniously kicked out of the application, losing any unsaved data.
It took us a while to figure out the problem, and as usual, about 30 seconds to fix it once we figured it out. A nice "hoisted by my own petard" problem to solve on a Friday. Ugh.
What I should have done, though, was edit this policy so that the mapping was set to "Update" after the migration was finished. Because I missed this step, we chased a problem for over a week with one of their LOB softwares kicking out users if they left the software up for an extended period of time. What was happening was that the "replace" policy was running (the schedule is different for every user since they don't all log into their computers at the same time.) When the policy ran, it momentarily disconnected the drives, then reconnected them, which made the LOB software think the network went down, so it crashed and the user was unceremoniously kicked out of the application, losing any unsaved data.
It took us a while to figure out the problem, and as usual, about 30 seconds to fix it once we figured it out. A nice "hoisted by my own petard" problem to solve on a Friday. Ugh.