> -----Original Message----- > From: [EMAIL PROTECTED] [mailto:nagios-users- > [EMAIL PROTECTED] On Behalf Of sunchai (gillnaldo : sun) > Sent: Thursday, September 20, 2007 8:08 AM > To: [email protected] > Cc: [EMAIL PROTECTED] > Subject: [Nagios-users] Nagios latency on service more than 1000 > > Hi nagios list, > > I've problem with service/host check latency with service more than 1000 > and host more than 100. Please advise me to tune-up as my requirement. > My requirement need service check & alert not more than 5 minutes. All > alert must check interval 3 times before alert. The server cpu > usage/memory use only 50% only.
Hmm. That implies normal_check_interval 2, retry_check_interval 1, max_check_attempts 3. You'll be doing 8 concurrent checks per second if they all exit immediately. You'll want to reap those service results as quickly as possible. Have you read the performance tuning documentation yet? There's a lot of good stuff there. > The nagios summary show on > ======================================= > HOST SCHEDULING INFORMATION > --------------------------- > Total hosts: 170 > Total scheduled hosts: 170 You certainly don't want scheduled host checks. With nagios-2 and prior, everything else stops while a host is being checked. If every host check completed in 1 second you've just told nagios to not check services, or do anything else for that matter, for about 3 minutes out of every 5. Nagios will automagically figure out if it needs to run a host check as necessary based on the state of its services. -- Marc ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ 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
