mirror of
https://pagure.io/fedora-infra/ansible.git
synced 2026-03-20 03:57:02 +08:00
22153994ba
Nagios warns us if status.fp.o isn't running, that's the goal. But nagios was checking for the presence of the string: "All systems go". This is fine, until one system goes down. Nagios tells us about this system, we go look at it, we (manually) update status.fp.o so our users know that we know about the outage. Then nagios tells us that status.fp.o isn't how it should be and we need to go tell nagios that we know status isn't how it should be since we updated it ourselves. So instead of checking for "All systems go" we'll now check for "Fedora Infrastructure Status" which is at the top of the page and will remain there as long as status.fp.o is up and regardless of the state of the rest of the infra. Signed-off-by: Pierre-Yves Chibon <pingou@pingoured.fr>
===================================
Nagios 4 Configuration for Fedora
===================================
The Fedora Infrastructure Nagios is built on a set of configurations
originally written for Nagios 2 and then upgraded over time to Nagios
3 and then 4.08. With additional changes made in the 4.2 series of
Nagios this needed a better rewrite as various parts came from
pre-puppet and then various puppet modules added on top.
In order to get this rewrite done, we will use as much of the original
layout of the Fedora ansible nagios module but with rewrites to better
match current Nagios configurations so that it can be maintained.
Role directory layout
=====================
The original layout branched out from
roles/nagios/client/
roles/nagios/server/
With the usual trees below this. This breaks ansible best practices
and how most new modules are set up so the rewrite uses:
roles/nagios_client/
roles/nagios_server/
=====================
Nagios Server Files
=====================
The Nagios Server Files require a large layout change. The original
Nagios system used multiple independant modes and files which caused
problems when hosts were removed. The new system will use hosts set up
from the Fedora Ansible Inventory with hostgroups set up to match
groups.
roles/nagios_server/{files,handlers,tasks,templates}
r.../n.../files/httpd ==> /etc/httpd/conf.d files
r.../n.../files/nagios ==> /etc/nagios/ files
r.../n.../files/nagios/commands command files
r.../n.../files/nagios/hosts host files
r.../n.../files/nagios/hostgroups groups made from hosts
r.../n.../files/nagios/services services
r.../n.../files/nagios/servicegroups groups made from services
r.../n.../files/nagios/contacts files for people
r.../n.../files/nagios/contactgroups groups made from contacts
similar layout for templates
handlers has the ways to restart and check configuration
tasks has the main rules for building stuff.
===================
Nagios Module Steps
===================
1. Check to see if the nagios user is configured. Someone years ago
chose that our monitoring uses UID/GID 420. Har Har.
Setup any other groups and permissions
2. Install the needed packages for the server.
3. Setup the directories on the server
/etc/nagios/{child}
4. Synchonise over the static files
/etc/nagios/commands/
/etc/nagios/services/
/etc/nagios/servicegroups/
/etc/nagios/contacts/
/etc/nagios/contactgroups/
/usr/lib64/nagios/plugins/
/usr/local/bin
/usr/share/nagios/html/
5. Build template files
/etc/nagios/commands/
/etc/nagios/hosts/{ansible-inventory, ansible-vars, other}
/etc/nagios/hostgroups/
6. Fix selinux policy
7. Restart services