Currently supported components:

• faf
• retrace-server

Using Vagrant

Simplest way to get instances of abrt stack running is:

• git clone https://github.com/abrt/abrt-ansible.git
• cd abrt-ansible
• ./fetch_remote_roles
• vagrant up

This will create virtual machines using libvirt backend and provision them using Ansible in similar way as described in the next section.

By default vagrant up will create multiple machines with abrt service running inside. You can also specify instance to run and provision:

• vagrant up faf
• vagrant up rs

Provisioning with Ansible

Following steps can be used to deploy and configure faf instance:

• create a virtual machine (either Fedora or EL/CentOS)
• edit inventories/hosts to use hostname or IP address of your testing machine
• use ssh-copy-id to copy correct ssh key to the testing machine's root account (there's a private_key_file configuration option in conf/config which specifies a private key file ansible uses)
• point your browser to the hostname or IP address of your testing machine to verify the installation
• proceed with further configuration of your new instance, for all available configuration options see group_vars/all file.
• it's recommended to use host_vars/<domain> to set configuration options for your machines.

Running database migrations:

• ./run --extra-vars="faf_migrate_db=true"

Retrace server

Retrace server provisioning is similar to previous section except instead of passing faf_first_time setup to extra-vars we have to pass selinux=false as retrace server doesn't support SELinux.

• edit inventories/hosts and add your machine to [retrace_server] group
• ensure ssh access via key used by Ansible
• run ./run --extra-vars="selinux=false"

Notes

./run wrapper passes parameters to ansible-playbook so you can add -vvv for debugging or limit host selection with:

-l SUBSET, --limit=SUBSET
           Further limits the selected host/group patterns.