RE: [jffnms-users] Poller2 Questions

[Broeken, Marco]

top - 16:36:53 up 2 days,  4:16,  1 user,  load average: 0.34, 0.36, 0.37 Tasks:  84 total,   1 running,  83 sleeping,   0 stopped,   0 zombie Cpu(s):   3.5% user,   5.5% system,   0.0% nice,  91.0% idle Mem:    513520k total,   499752k used,    13768k free,    89664k buffers Swap:  4096552k total,        0k used,  4096552k free,   235216k cached we have 44 hosts and 441 interfaces   we use a virtual server with 512 Mb ram, so performace would be bad compared to your setup...   we use 2 pollers to minimize the load on our setup # New Poller for big Networks */1  * * * *    jffnms  cd /home/global/www/jffnms/engine && /usr/bin/php -q poller2.php master 2 > /dev/null 2>&1   we noticed a difference on 2 or 5 poller setups   on diskspace we use a harddisk of 30 Gb, this is now expanded to 2 Gb, and the jffnms cronjobs are cleaning it every week   Met vriendelijke groet,   Marco Broeken Netwerkbeheer VECOZO BV Van: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Namens Kory Verzonden: woensdag 17 mei 2006 16:33 Aan: jffnms-users@lists.sourceforge.net Onderwerp: [jffnms-users] Poller2 Questions I’ve been watching the load level on my polling machine. (I have a separate Database/Web server, this machine only handles the polling processes). The polling machine is a dual 2 GHz Xeon with 1 Gig Ram and 4x 74Gig 10,000 RPM Raptor Drives in a Raid 10 configuration.  I’m monitoring 46 hosts and 1039 interfaces and my processor load level is at .78 on a 5 minute average, .60 on a 15 minute average.   My first question is:  Is this normal for this type of server to be running this hot?  I have another 50 - 100 hosts (1000-2000 interface) that I want to monitor and I’m not sure if it’s going to be able to take the load.   My next question is about the Poller2 process.  What is normal operation for this process.  I’m seeing a master poller process running and then every minute it looks like another starts and then exits. The child poller processes never seem to exit either so I can’t tell if the polling actually completes every 5 minutes or if that’s the reason for the high load level. Is there a way to tell if the poller made it’s rounds within 5 minutes?   Also, before I split up the Poller and Database/Apache machines I did some figuring based on the performance of the poller on a dual 733 MHz and figured that I could Monitor ~200 hosts & ~4000 Interfaces with a dual 2.0 GHz machine with 150 GB of Hard drive space.  Do you think this is accurate?   Thanks in advance, Kory Knopp   -- No virus found in this outgoing message. Checked by AVG Free Edition. Version: 7.1.392 / Virus Database: 268.6.0/341 - Release Date: 5/16/2006