Looks like XI is dropping Passive Checks? Tuning Question
Posted: Thu Dec 23, 2010 1:38 pm
I use NSCA, I have that Cron'ed to send back data every 5 minutes and it does so happily. Services have a 900 sec Freshness threshold. Nagios has been going bonkers lately because it seems like Nagios is not able to keep up with processing all the checks and is either constantly trying to clear a queue or dropping things on the floor... This is causing a lot of false positive alerts when Freshness checking kicks in and thinks hundreds of services have gone CRIT at the same time.
How can I tweak out XI to process these requests quicker I need all my checks per host to come in at the 5min mark... there should never be a service or host check that takes over 15 unless there is a legit problem. My command_check_interval is set to "-1" which AFAIK should mean Nagios will process everything immediately. Resources are generally happy on the VM with load sitting around "2' and Mem at about 1GB free for both Physical (virtual) and Swap.
How can I tune this? is there a Tuning guide for hight performance?
Thanks
How can I tweak out XI to process these requests quicker I need all my checks per host to come in at the 5min mark... there should never be a service or host check that takes over 15 unless there is a legit problem. My command_check_interval is set to "-1" which AFAIK should mean Nagios will process everything immediately. Resources are generally happy on the VM with load sitting around "2' and Mem at about 1GB free for both Physical (virtual) and Swap.
How can I tune this? is there a Tuning guide for hight performance?
Thanks