kbidarkar/satellite-monitoring

Monitor Red Hat satelitte & capsule servers

Satellite-Monitoring

Satellite-monitoring contains simple Ansible playbooks that can be used to perform following tasks:

Monitor Satellite & Capsule Details

   *System Resources*
        - cpu, Memory, Disk, Network, Turbostat, Numa, IRQ, Load time
   *Satellite processes Memory, cpu, pagefaults, IOPS, IO throuhput*
        - Dynflow, Forman-smartproxy, Mongodb, Passenger, Postgres, Pulp, Puppet-agent, Qpidd, Qpid Dispatch Router, Tomcat,  
   * Satellite Database Operations*
        - Foreman, Candlepin

Table of Contents

• Introduction
• Preparing your system for setup
• Setting up the hosts file
• Creating your own personalized configuration for monitoring
• Deploying your monitoring setup
• Installing optional statsd module

Introduction

Satellite-monitoring package provides a way through which Satellite 6 users can install Monitoring tools for their Satellite and Capsule installation.

The following setup is currently provided by the satellite-monitoring:

• collectd
• statsd
• grafana
• graphite If you want to setup a monitoring for your Satellite installation, follow on with the guide.

Getting Started

Ideally, you need Three hosts to run this project:

1. Ansible Control node (referred to as Control node in the rest of this document) is the host from which satellite-monitoring ansible playbooks are run. Could be user laptop
2. Destination node - Ansible playbooks install collectd on below nodes
   • A Satellite server
   • Capsule servers
3. Monitoring server - Ansible playbooks install below services
   • Grafana
   • Graphite
   • Carbon

Note

1. You can get away with using one host by optionally choosing to use Destination node as the Control node.
2. Make sure that the Control node can connect to the Destination node & Monitirng Server via paswordless ssh.

On the Control node:

Supported versions

• RHEL 6
• RHEL 7

1. git clone this project.

     # git clone https://github.com/redhat-performance/satellite-monitoring.git

   NOTE: Optionally you may utilize the script [control_node_setup.sh] (adhoc-scripts/control_node_setup.sh) to perform step 2 below. The instructions to use this script are documented in the script itself.

2. Install ansible package on the Control node. For RHEL boxes, [access to EPEL] (https://access.redhat.com/solutions/3358) is required.

     # yum install -y ansible

3. Create an inventory file named inventory (by copying ansible/hosts.ini.sample) and update satellite, capsule, grafana, graphite it as necessary:

  # cp conf/hosts.ini.sample conf/hosts.ini

To run with raw ansible-playbook commands on commandline

To install Collectd on satellite & capsules:

$ ansible-playbook --private-key conf/id_rsa -i conf/hosts.ini ansible/collectd-generic.yaml --tags "satellite6"

$ ansible-playbook --private-key conf/id_rsa -i conf/hosts.ini ansible/collectd-generic.yaml --tags "capsules"

To install graphite server, grafana on Monitoring server:

$ ansible-playbook --private-key conf/id_rsa -i conf/hosts.ini ansible/grafana.yaml

$ ansible-playbook --private-key conf/id_rsa -i conf/hosts.ini ansible/graphite.yaml

To build graphite dashboards:

$ ansible-playbook --private-key conf/id_rsa -i conf/hosts.ini ansible/dashboard-generic.yaml

If collectd fails to send metrics to your grafana instance

You might wanna check the selinux policies. Try one of the following to counter "Permission Denied" log statement:

# setsebool -P collectd_tcp_network_connect 1

OR

# audit2allow -a
# audit2allow -a -M collectd_t
# semodule -i collectd_t.pp

OR

# semanage permissive -a httpd_t

..or all.

To view your Dashboard visit:

http://:11202