Though there will definitely be no posts for a at least a month.

Happily, it is a painless gem install. See: http://www.stylesen.org/openshift_client_tools_installation_debianubuntu

There are lots of how-tos already available. The deployment was very quick, and painless. While I do not feel the need to copy the howto verbatim I will include a couple of things unique to WordPress and Openshift.

Openshift does not provide a static IP address to your app.

• You will need to setup a CNAME on your DNS server to [appname]-[namespace].rhcloud.com.
• You then need to add the alias to openshift
  $ rhc alias add [appname] [alias]
• WordPress also needs to be configured with the correct hostname or every link will direct you to [appname]-[namespace].rhcloud.com. Under General setting you can change the hostname and siteurl: 

Your app is deployed via GIT. All the usual buttons work in WordPress for installing themes and plugins, but if you pull from you git repository, or push from a local copy all of these changes will be overwritten.

• Make changes to the GIT repository and push rather than make changes through the WordPress admin interface. Themes and plugins can simply be placed in their respective directories under “php/wp-content”.
• Or if you do make changes on the production site, you can rsync the changes from your openshift deployment back to your git clone and add the changes there.
• If you have changed the WordPress permalink settings there will be a “.htaccess” file created on the Openshift gear. Grab this and add it to your git repository or, every time you push your changes, the permalink settings will be reset.

But it is fairly straight forward to get ADB, if your using the Developer Tools, and MTP mounting working.

1. Create the file /etc/udev/rules.d/99-android.rules
   
# Google Nexus 7 16 Gb
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e41", MODE="0666", OWNER="your-login" # MTP media (multimedia device)
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e42", MODE="0666", OWNER="your-login" # MTP media with USB debug on(multimedia device)
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e43", MODE="0666", OWNER="your-login" # PTP media (camera)
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e44", MODE="0666", OWNER="your-login" # PTP media with USB debug on (camera)
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e40", MODE="0666", OWNER="your-login" # Bootloader
SUBSYSTEM=="usb", ATTR{idVendor}=="18d1", ATTR{idProduct}=="d001", MODE="0666", OWNER="your-login" # Recovery
# MTP Support
ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e41", SYMLINK+="libmtp-%k", ENV{ID_MTP_DEVICE}="1", ENV{ID_MEDIA_PLAYER}="1"
ATTR{idVendor}=="18d1", ATTR{idProduct}=="4e42", SYMLINK+="libmtp-%k", ENV{ID_MTP_DEVICE}="1", ENV{ID_MEDIA_PLAYER}="1"


2. Restart udev
   
# systemctl restart udev.service


3. Install required software
   
# yum install mtpfs libmtp


Done.
You should now be able to mount your device:

$ mtpfs /home/user/some_mount_point


The existing help I have found on MTP failed to include both Product IDs. Meaning if USB debugging was enabled on the device libmtp would not recognise your device.

I set this up on 2 KVM guests using the default network configuration. In the examples I will use the ip addresses of my guests, change them to match your setup.

1. Prerequisites

Software Packages

You will need the EPEL repository setup.

# yum install http://mirror.iprimus.com.au/epel/6/i386/epel-release-6-7.noarch.rpm

Install the required packages on both machines.

# yum install corosync pacemaker openvpn

You should set corosync to run on startup. But do not set openvpn to run. Pacemaker will start OpenVPN when required.

# chkconfig corosync on

Firewall

If your lazy you can disable the firewall but I would suggest opening the required ports. For corosync, use the following filtering. Port 5405 is used to receive multicast traffic.

iptables -I INPUT -p udp -m state --state NEW -m multiport --dports 5404,5405 -j ACCEPT

If you are installing OpenVPN allow udp traffic on 1194.

2. Configure Corsync

On our first host.

Setup the corosync authkey

# corosync-keygen
Press keys on your keyboard to generate entropy.
Press keys on your keyboard to generate entropy (bits = 160).

Note. If you are remotely logged in, no amount of key bashing will work. You need to be connected to console for key-smashing to result in entropy. One novel suggestion for generating IO I found on a forum, was to download the latest kernel source, untar and then run a find over the resulting directory tree.

Copy the file “/etc/corosync/authkey” to the other host. Set the permission to “0400″.

corosync.conf

We will use the example config included. Copy this file and edit it.

cp /etc/corosync/corosync.conf.example /etc/corosync/corosync.conf

We need to set bindnetaddr to our local subnet:

bindnetaddr: 192.168.122.0

And we need corosync to start pacemaker. Add the following to the end of the file

service {
	# Load the Pacemaker Cluster Resource Manager
	name: pacemaker
	ver: 0
}

Start Corosync

# service corosync start

Once corosync has started successfully on the first host:

# ssh 192.168.122.21 -- service corosync start

Run “crm_mon” you should see that corosync is running on 2 nodes.

Setting up Active/Passive Cluster

Disable Stonith:

crm configure property stonith-enabled=false

Add a cluster IP

crm configure primitive ClusterIP ocf:heartbeat:IPaddr2 params ip=172.25.3.20 cidr_netmask=21 op monitor interval=30s

Disable Failback (there is no need to failback in this case)

crm configure rsc_defaults resource-stickiness=100

With 2 nodes we cannot attain a quorum

crm configure property no-quorum-policy=ignore

Print the current config

crm configure show

You should now be able to ping you new cluster IP address 192.168.122.23. This ip address should remain available if you turn off corosync on the first host, or even shut it down.

Adding a service

Configure OpenVPN

This is not a post about setting up openvpn. I wrote about that previously, though the post is specific to Debian. I have included my config file though.

Once configured copy the entire directory structure “/etc/openvppn” to the second host.

$ cat /etc/openvpn/openvpn.conf

# This is the clustered IP
local 192.168.122.23

dev tun
proto udp 
port 1194 

ca /etc/openvpn/easy-rsa/2.0/keys/ca.crt
cert /etc/openvpn/easy-rsa/2.0/keys/centosvm.crt
key /etc/openvpn/easy-rsa/2.0/keys/centosvm.key
dh /etc/openvpn/easy-rsa/2.0/keys/dh1024.pem

user openvpn
group openvpn
## You can make this any private subnet you like
server 192.168.123.0 255.255.255.0

persist-key
persist-tun

client-to-client

## make this connection the default gateway for network traffic
push "redirect-gateway def1"
push "dhcp-option DNS 192.168.122.1"
    
log-append /var/log/openvpn

plugin /usr/lib64/openvpn/plugin/lib/openvpn-auth-pam.so system-auth
client-cert-not-required
username-as-common-name

Add Pacemaker rules

# crm configure primitive p_openvpn ocf:heartbeat:anything params binfile="/usr/sbin/openvpn" cmdline_options="--writepid /var/run/openvpn.pid --config /etc/openvpn/openvpn.conf --cd /etc/openvpn --daemon" pidfile="/var/run/openvpn.pid" op start timeout="20" op stop timeout="30" op monitor interval="20"
# crm configure commit

If all of this has worked you should now have an Active/Passive OpenVPN cluster.

The following sssd.conf worked for our environment. (Making it work with FreeIPA or Open LDAP and Kerberos is far less fiddly).

[sssd]
domains = MYDOMAIN
services = nss, pam
config_file_version = 2
sbus_timeout = 30
 
[pam]
offline_credentials_expiration = 0
 
[domain/MYDOMAIN]
description = MYDOMAIN AD Server
#debug_level = 9
enumerate = true
ldap_referrals = false
min_id = 1000
 
access_provider = permit
 
id_provider = ldap
chpass_provider = krb5
 
ldap_uri = ldap://my.ldap.server
ldap_search_base = dc=my,dc=ad,dc=domain
 
ldap_id_use_start_tls = False # TLS/SSL is supported

# If you do not have anonymous binds enabled 
# User that can read from AD, any normal user should work. Update as necessary
ldap_default_bind_dn = user@my.ad.domain
# Leave this as password
ldap_default_authtok_type = password
 
# The ldap users actual password, update as necessary
ldap_default_authtok = password

# This is the important stuff for making AD LDAP work 
ldap_schema = rfc2307bis
ldap_user_principal = userPrincipalName
ldap_user_fullname = displayName
ldap_user_name = sAMAccountName
ldap_user_object_class = user
ldap_user_home_directory = unixHomeDirectory
ldap_user_shell = loginShell
ldap_user_principal = userPrincipalName
ldap_group_object_class = group
ldap_force_upper_case_realm = True

# I love this setting
override_homedir = /home/%u
 
# kerberos config
auth_provider = krb5
krb5_server = dc.my.ad.domain
krb5_realm = MY.AD.DOMAIN
# This will probably not work for changing passwords
krb5_changepw_principle = kadmin/changepw
krb5_ccachedir = /tmp
krb5_ccname_template = FILE:%d/krb5cc_%U_XXXXXX
krb5_auth_timeout = 15
cache_credentials = True

To fix add the grub option:

"reboot=pci"

The correct way to do this is to edit /etc/default/grub and then issue:

grub2-mkconfig > /boot/grub2/grub.cfg

“exim -bt email@address” is reason enough

It is simpler to setup and easier to administer. If you want your fedora machine to send local email via Gmail, here is a guide.

One point to add, if you want root mail delivered somewhere, edit /etc/aliases and set an alias for root. Don’t forget to run “newaliases” when your done.

It is pretty simple though, you just need a boot option:

radeon.audio=1

If this is passed in via grub at boot time, audio will work.

Footnote 9 on this page: http://www.x.org/wiki/RadeonFeature

This config change will get the applet to launch Thunderbird (or any  other app), but it will still not populate the calendar with entries from Thunderbird. But this single line, will mean that Thunderbird launches when you click on the in-built calendar.

$ gsettings set org.gnome.desktop.default-applications.office.calendar 
exec 'thunderbird'

To populate the calendar widget there is an extension for Thunderbird called: Evolution Mirror which should populate the evolution db with your Thunderbird data. You will need to install a couple of additional packages to make it work, in F16:

$ sudo yum install gnome-python2-extras gnome-python2-evolution