EZ can you clarify. Is this a remote console on some device other than the core, or did you install the agent onto the core itself?
If this is just a remote console on a device other than the core, I don't see why you should be having any issues.
If you installed the agent onto the core itself, you probably have to .... call support and verify....
now that you were warned, I believe you need to uinstall it while leaving the CBA stuff. The command to do this is;
For years they have warned not to install the agent on the core server, I still don't understand why there isn't a block in the installer that detects if its a core, not to proceed with the install... or maybe they have done this already, its not like I've tried to install it on the core or want to test an see, but if not, they should (queue someone telling me to go post it to the ideas board).
We have installed the agent on our cores for the past 10 years without an issue. Doing an agent uninstall with the forceclean option is a bad thing on the core.
I have been back and forth with LANDesk on this... yes, it is fine... no don't do it... what day is it?? I think it depends on who you to talk to and whether it's something that has been passed down and may not be true, or if there is a legit reason on why not.
I think it is pretty sad that if there is a reason why not to, that there is a reason at all.
From a more historical perspective...
- Back in the olden days (early 2000-s ... so around LDMS 6.6x / 7.0 / 8.0 versions) it was a bad idea for sure to install the agent on the Core.
- Chief reason for this was that as part of the uninstall, we'd be removing a whole slew of registry keys that the Core needed to actually function properly (was fine for an agent to not have them -- but the Core would break).
- There *WAS* some other reason why installing the agent was bad back in the day, but I can't recall what it was (and seeing as it's 15+ years in the past, I'm loathe to search through e-mail archives & such) ... but there were other reasons back then as well.
- This was all improved around the 9.0 era I think (it's been a while for sure), when the uninstaller was changed to not remove a "said bunch of things" by default ... so you could have an agent on the Core and not risk breaking it. So "yay".
... since then, you can have a regular agent on the Core.
- AV on the Core could be a bit more of a headache (mainly due to config & needing to configure the AV to not tank disk I/O) ... but that's "the same story" whichever AV is/was used, not specific to what we integrated with. People can (and do) still kill their cores with a "default configured" AV without consideration for what we do & how we do it.
The big "No-no" in the past was UNINSTALLING the agent more than anything (because it used to break the Core). That's now a more acceptable thing (and I've been upgrading Cores for the last 10-odd years by starting to remove the agent from them) ... but there's a lot of "old wives' tales" type of thinking in the support org (we get burned a few times, things get put into stone, even if the rules change) among the support org, with a few people still having been around in the day to remember how that whole thing came about.
So - long story short.
- Installing an agent on a Core is fine.
- You need to configure your AV to behave sensibly on the Core anyway (not an item in this subject, just covering my bases).
- Where things USED to go awry was UN-installing the agent (it used to wipe out te DeviceID and removed critical files for the Core's own functioning which were shared among other things, which led to "broken Cores"). But this hasn't been an issue now for 10+ years if not longer.
- Console & Agent are disconnected ... you can (/should) happily have a console on a device with or without an agent. Shouldn't affect each other.
- If you have problems with the console ("it doesn't work" isn't exactly a detailed description), start with the logs first of all (CONSOLE.EXE.LOG for a start) and see where that leads you. Push comes to shove, tag support to help you out.
Installing the agent shouldn't (in my experience - doesn't) break the Console. 2 separate sets of files & such ... the two shouldn't touch in any significant way.
Hope that helps?
Sorry was away for a while. Now i am back. I have not actioned this yet. Still open to suggestions.
Rick.Smith1, this happened on 2 use cases:
1) My work laptop already had an agent and i tried to install the console so i can remotely administor without RDP into the EPm core.
2) The core Server was added to the deployment task (by accident). The console was working but post agent deployment it would not launch.
The use case that is most important to me is point one. i already have an agent and my laptop is a managed device. I want to also install the console. Any ideas?
PS: Thank you phoffmann for the background info - very useful and i will definitel;y NOT uninstall to prevent further damage.