Hello Carl,
> Every end-node? Why? I can see monitoring servers, or if this is some > odd form of server-farm I can see it, but for plain old desktop systems? Consider IoT: the internet of things. We have here 10.000's of specialized devices connected to a network, which we want to monitor for network connectivity. If one fails, simply replace it by a new one. Thanks for the help, best regards Carl. -- Carl Wolff Consultant Imtech ICT Technical Systems A: Basicweg 16, 3821 BR Amersfoort T: +31 (0) 33 4500412 M: +31 (0) 06 53469668 E: mailto:carl.wo...@imtech.nl W: http://imtech.eu/NL/ICT/Technical-Systems L: http://www.linkedin.com/in/carlwolff ________________________________________ Van: Carl R. Friend [crfri...@rcn.com] Verzonden: vrijdag 11 oktober 2013 14:11 Aan: icinga-users@lists.sourceforge.net Onderwerp: Re: [icinga-users] Scale of network On Thu, 10 Oct 2013, Carl Wolff wrote: > We want to monitor every(! )node (including end-nodes) on availability, > that is a total of (8*750)+750+48+1 = 6799 nodes. Every end-node? Why? I can see monitoring servers, or if this is some odd form of server-farm I can see it, but for plain old desktop systems? In any event... > If I enter this in icinga, i end up with a very slow performing status > map, filled with circles cluttered with end nodes who's description > cannot be read... Part of the problem is that the default layout (circular with markup) doesn't scale well with lots of hosts involved -- even if one is very diligent with setting up the parentage rules. The "Balanced tree" or "collapsed tree" maps are *better* in this regard, but with the number of hosts you're dealing with those, too will be congested. > How can above problem be mitigated? Do I need more hierarchy? It's usually best to let the network design define the hierarchy as that'll allow the parent/child relationship to identify trouble- spots more accurately. Specifically, this'll allow Icinga to call out switch or comms failures instead of bombarding you with "host down" notifications instead of pinging you for a powered-off switch. For a better solution, if a map is what you need, you could use NagVis which will allow you to break the map up into sub-maps and then rotate those maps around in sequence. Cheers! +------------------------------------------------+---------------------+ | Carl Richard Friend (UNIX Sysadmin) | West Boylston | | Minicomputer Collector / Enthusiast | Massachusetts, USA | | mailto:crfri...@rcn.com +---------------------+ | http://users.rcn.com/crfriend/museum | ICBM: 42:22N 71:47W | +------------------------------------------------+---------------------+ ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk _______________________________________________ icinga-users mailing list icinga-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/icinga-users ------------------------------------------------------------------------------ October Webinars: Code for Performance Free Intel webinars can help you accelerate application performance. Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from the latest Intel processors and coprocessors. See abstracts and register > http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk _______________________________________________ icinga-users mailing list icinga-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/icinga-users
