3 Replies Latest reply on Feb 11, 2019 11:12 AM by rschroeder

    How to get live infrastrucutre of network along with the servers and PC

    cloudev

      I'm looking for live network and server infrastrucutre status , which gives the live update of the PC's and other miscellaneous devices connected in the network . How do i achieve this in solarwinds ?

        • Re: How to get live infrastrucutre of network along with the servers and PC
          rschroeder

          There are a few different ways you might accomplish this, and I'm betting other Thwack people or SolarWinds staff have even better ideas than mine.

           

          • Monitor every node with NPM.  Routers, switches, servers, PC's AP's, printers, etc.  Everything.  NPM will let you know when it's up and when it's down.  This is what NPM does--but you've taken it to an extreme degree by monitoring EVERYTHING.  Usually we don't care whether a PC is up or down.   This particular solution will use the most resources, the most amount of licenses, and generate the most alerts.  Do you really need to know when a PC is offline?  If so, OK, this will do it.  If not, why monitor & alert on the status of a PC--especially if there's nothing you can do about it.  We want actionable alerts only, not alerts that cause alert fatigue or that fill you e-mail with spam that you don't care about.  I don't see this as a good solution for an enterprise, but the smaller the shop, the more easily it is handled by NPM.
          • Monitor only important nodes.  Not PC's, not printers, etc.  This will reduce the alert fatigue, but won't get you status of PC's or other miscellaneous devices on the network (which is your original request).
          • Deploy a SIEM like LEM or Splunk and ensure all your nodes report everything to that SIEM.  You'll see port up and port down info, but you'll have to parse through it all to create useful alerts.
          1 of 1 people found this helpful
            • Re: How to get live infrastrucutre of network along with the servers and PC
              cloudev

              My goal is to get visualized live status of the infrastrucuture as i have mentioned the PC's those are present in the remote locations , so we require to check the machine status is up or down . Below is the example what i'm looking to achieve. I want to know visually when something goes down and trouble shoot like alright is the router is up - check , Switch is up - check , alright then something wrong in the PC side.

               

              Network Mapping Software | SolarWinds was checking about this , but does it show us the live status of the infra ?

              Image result for softinventive

                • Re: How to get live infrastrucutre of network along with the servers and PC
                  rschroeder

                  The maps I build with Solarwinds tools DO show node status as red for down and green for up.

                   

                  Typically, here in the Network Team, we prioritize our monitoring in this order:

                  1. Data Center and Core and WAN resources (including firewalls, Internet, DMZ's, ACI, etc.)
                  2. Distribution Switches and Border Routers
                  3. Access Switches
                  4. AP's
                  5. Nodes

                   

                  We do not manage or monitor servers; our System Administration team is responsible for knowing if they are up or down.  Similarly, we do not monitor PC's, printers, or other end nodes.  Their users are responsible for that, and they can easily tell if a device has lost power or lost network link light.  We train them for this and trust them to understand that six Network Analysts triage outages based on scope.

                   

                  If a single PC is offline and a Border Router or access switch above it are also offline, we won't spend time on the PC.  We'll go work down from the top of that list and troubleshoot the first thing that's unavailable--because it affects the most people.  We get the most return for our efforts by restoring a Distribution Switch or Border router than we do by troubleshooting the individual PC that may be down--BECAUSE the router or distribution switch above it is preventing access by that PC.

                   

                  Once everything is up except for a node, my team still isn't the first group to work at the node level to restore connectivity to it.  We have End User Platform Support people who troubleshoot at the PC level, verifying things like power, basic network connectivity, etc.  After they verify all that is good for a PC and it still doesn't talk to the network, they escalate the issue to my team and we look at logs and port configurations to verify STP is working as expected, that ISE has't shut down a specific device's access, that port-security hasn't been violated, etc.

                   

                  I think in your scenario you may wish to pursue a Network Map with a layered hierarchy similar to the five groups above. I've done that in the past and it's worked out fairly well

                   

                  Top Group: ALL Network.  It turns yellow if something below it is down.

                  Sub group 1:  All individual sites, including Data Centers.  Individual sites turn yellow if something within the site is down, or turn red if all things withing the site are down.

                  Sub group 2:  Network Rooms.  A site may have twenty network rooms.  Each gets their own drawing, and we build those drawings on imported floor plans that have the network nodes dragged onto them and positioned inside them accurately.

                   

                  From a high level view we can just leave the Top Group up.  If it's green all is well.  If it turns yellow we can double click on it and see which sites have problems, and drill into them further and further.

                   

                  NPM does this natively, although not with maps.  Given I have a hundred sites and several hundred network rooms, it's not practical to map and track status of the 50,000+ end devices out there.  I monitor the AP's and switches and routers and know that if one of them is down it affects multiple users.  And I allocate my support time to fix them in order from highest impact / most affected users to lowest.

                   

                  If you are in a much smaller shop, you can determine if it's worth your time to track individual devices like PC's.  You may run into challenges with tracking individual devices whose IP addrresses change as they get new leases from DHCP and as they move from location to location, or as they are retired and replaced by newer devices.

                   

                  Don't bite off more than you can chew.

                  1 of 1 people found this helpful