3 Replies Latest reply on Dec 7, 2017 5:17 AM by phoffmann

    Issues with Self electing services in Console

    RobLent Specialist

      I can't see any services under the above tab in the console.


      I have found this document which describes the issue we are having but the steps do not fix my issue:  Unable to enable ANY service under "Self-electing subnet services"


      I have found that the IIS Admin service will not start on the Core server which I am guessing may be the problem for us.


      Looking at the errors and doing some research it seems that a previous issue we have had on the server may be casuing this problem.


      We had to raise a call with Support as we could not login to the console and it was found that there was a problem with the IIS certificate.  The support engineer recreated a new certificate which got us working again but since then the IIS admin service will not start.


      Solutions on the web state to uninstall IIS and re-install it.


      Can I do this on the Core server without breaking anything more?

        • 1. Re: Issues with Self electing services in Console
          phoffmann SupportEmployee

          So "uninstalling IIS & re-installing it" is usually not a good idea - at that point, you may essentially start with a fresh Core.


          It'd probably a solution if you're "just hosting a few static www-pages / a directory structure" but since we're running entire app-pools with specific identities & so on, that strikes me as very dangerous advice to take.


          *IF* you're looking at this, the very first step to do before even thinking about uninstalling IIS is to make a FULL backup of your Core. Your Core may have issues with CSEP at the moment, but if you break IIS (as far as we're concerned), then you pull out the rug from about 70-80% of "how we operate" and have essentially a Core Server that may not even do inventory (as that's also done via IIS to a degree). So this is a HIGH RISK operation to undertake!


          Then you can experiment with saving the IIS configuration (as it presently is), uninstalling & re-installing the IIS components and re-loading the config and re-registering .NET ... and hoping for the best (IIS has a LOT of finickey config items, any single one of which can mess you up).


          Personally ... especially since you mention "the server had previous issues", my advice is that with any "unknown system" is to "look to start fresh". This has the benefit that this doesn't break any existing stuff & gives you great control all the way as you set the pace of migration..


          Start with a new OS install, a new LANDesk install ... and migrate things (and clients) over. That'll take about 1/2 a day to a full day (-ish - may vary depending on a few complexity factors), but is usually much faster than high risk experimentation (because "something" will go wrong inevitably, and you'll end up chasing your tail for a week just to find a single setting that's responsible  ... and then having to repeat the process with whatever's next causing you issues).


          This is - incidenally - also a great opportunity to up-lift both the Core Server OS & version of the management suite / EPM that you use.


          It sounds painful, but I've long developed a distrust & dislike towards "unknown / questionable state" servers ... because there's always SOMETHING up with them, and you never know whether things not working are because of some of that shenanigans.


          Hope that helps a bit?

          • 2. Re: Issues with Self electing services in Console
            RobLent Specialist

            Thanks phoffmann.


            I want to do the new server from fresh bit but have resistance to doing this here.


            Hopefully this information will help convince those that be that this is the way forward.


            Thanks again for the detailed response.

            • 3. Re: Issues with Self electing services in Console
              phoffmann SupportEmployee

              What sort of resistence are you encountering?


              If there's any messaging or "reasons why this should need doing", we're (well - I) am generally happy to help out. The reasons above are the technical reasons of why it makes sense ... sometimes it just takes the relevant "external provider" letterhead / e-mail address for management to accept that this is indeed necessary.


              So - you shouldn't feel all alone with that. We're generally happy to "try & do what it takes" to get the right thing done. It's an unfortunate trend more often than not that decisions get made by people woefully underqualified / under-informed to actually make them . (And - let's be honest - it's not like there's a massive cost here, as you're not buying a new, hot-spare super-computer ... once the migration is done, the old kit can be re-used).


              So it's usually a matter of throwing the numbers at management I find (ongoing cost of X, due to "stuff not working" ... which will steadily rise and will continue to be on-going as opposed to a one-off cost to "do things right") ... and lead them to the right well of knowledge to drink from (/accept what's common sense) .