On Mar 8, 2010, at 9:17 AM, HyunSung Jang wrote: > hello. > I am think about setting up distributed monitoring but i just wanna know > about how many > hosts or services can hold in one nagios server.
There's no simple answer to this question because there are too many variables. An equivalent question is 'How far will a car go on a liter of gas?" > I have about 1000 hosts in my environment, and 3-5 services for each hosts. > this makes maximum 5000 services check which i need to setup. Are these simple checks or complex checks? Are you performing them once a second or once every 24 hours? Are you checking things on a fast LAN or slow WAN? Answering these questions won't get you a definitive answer but does illustrate some of the factors that can affect how many things you can check in a given interval. > can I make one nagios server to check all this hosts and services? > and do you guys think that it is possible? I expect that if you have a mix of simple (like ping) and complex (like check_http) checks, performed every 5 minutes or so on modern hardware and a decent network, you should be fine. I have single machines checking ~4000 services with 5 minute interval (complex and simple checks) and I appear to have plenty of room to grow. > if not, how many hosts or services should hold in one nagios servers? > > at this moment i adjusted interval_length to 1 in nagios.cfg and i am using > 10 sec to check_interval in objects.cfg. I'd be surprised if you could run 5000 checks every 10 seconds, even if they were just ping (unless just a single ping). It may be possible but you'll have several thousand processes running concurrently all the time. > Any comments or recommends are appreciated. I think the best answer anyone can give you is Try And Find Out, given your environment, the checks you want to perform and the frequency you want to perform them. -- Marc ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Nagios-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
