Currently nagios is sending us way too many alerts (I know we should fix these but that's in the process of happening), in the mean time I need a way of ensuring our CORE hosts reach our email with a higher priority. The way i can envision this is being able to group the hosts into a "Core group" and somehow adding the hosts group or some sort of "Level" into the notifications...
How do you guys achieve this? Distinction between core hosts and other (for example development) host notifications that are not so important?
Distinction between important hosts and not so important
Distinction between important hosts and not so important
Nagios XI 2012R2.8c Running on Ubuntu 12.04 Using 99% passive checks for monitoring
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
Re: Distinction between important hosts and not so important
An idea that popped into my mind would be creating a new user/contact for each person to be alerted and call it "person-core" then adding all those contacts to a "core group" which would be assigned to the core Services/hosts and changing the alerting template via email to include "[CORE ELEMENT PROBLEM]" on the title of each email which could then be filtered in outlook to a specific folder... This is not scalable and seems more of a hackish solution, any other ideas?
Nagios XI 2012R2.8c Running on Ubuntu 12.04 Using 99% passive checks for monitoring
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
Re: Distinction between important hosts and not so important
Well, the best way would be to redesign your notification objects to make more sense. I usually suggest that people add minimal notifications at the start, and then just add what is necessary and the configuration progresses.
Your solution is not really a hack as the contact/group objects were designed to allow the flexibility your solution relies on.
Your solution is not really a hack as the contact/group objects were designed to allow the flexibility your solution relies on.
Former Nagios employee
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
"It is turtles. All. The. Way. Down. . . .and maybe an elephant or two."
VI VI VI - The editor of the Beast!
Come to the Dark Side.
Re: Distinction between important hosts and not so important
@abrist that is what we have done. We have added only what we need to be warned about and then kept adding and adding more machines...
After a certain point it becomes unmanageable and we are getting 300-400 notifications a day, hence some things need to be prioritized at the moment.
What i want is to be able to distinguish the alerts i get from my "Core Group" to be sure they are dealt with immediately while other alerts could be taken care of at a later stage and not so urgently.
We can't even enable SMS because of the amount of notifications we get...
Any more ideas how to achieve this ?
After a certain point it becomes unmanageable and we are getting 300-400 notifications a day, hence some things need to be prioritized at the moment.
What i want is to be able to distinguish the alerts i get from my "Core Group" to be sure they are dealt with immediately while other alerts could be taken care of at a later stage and not so urgently.
We can't even enable SMS because of the amount of notifications we get...
Any more ideas how to achieve this ?
Nagios XI 2012R2.8c Running on Ubuntu 12.04 Using 99% passive checks for monitoring
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
Monitoring nearly 800 Passive services spread through roughly 40 machines
Running on an 8 core, KVM virtualized VM, with 15 GB of RAM and using RAMDisk
-
slansing
- Posts: 7698
- Joined: Mon Apr 23, 2012 4:28 pm
- Location: Travelling through time and space...
Re: Distinction between important hosts and not so important
Well, what exactly to do, and how to order things would be up to you as the Nagios Admin. We can't really tell you what systems are more important than others and should be notified on at what rates. I would suggest you set up a contact group that is specifically for "Core" systems, and then a combination of contacts, and groups, for items that are not so important to you. Then adjust your notification options in those object's configurations so that the "not so important" objects notify over a greater period of time, or wait for their initial notification, and the core systems could possibly be set to notify immediately, and much more rapidly. Generally steps 4-5 of the wizards cover this rollout, but it can be done through the CCM as well.