21c4055364
* Implement a special puppet type to wait until haproxy backend is online. * Wait for mysqld backend as a requirement for nova and keystone backends * Switch to TCP connect for cluster check instead of UNIX socket: - the host to connect is the node's management IP - the port is the mysql HAProxy backend port, 3307 - the user is clustercheck and its password is the same as wsrep user has - the timeout is 10 seconds * Parametrise galera cluster check class and script * Move galera::cluster check to opesntack module (otherwise, there are 4 more classes to drag all the cluster check parameters through, including puppet-mysql module which does not have these params in upstream) * Fix docs for cluster check class parameters Related-Blueprint: pacemaker-improvements Closes-bug: #1394137 Change-Id: I4eab6af7257270bf3cb2b40a34bcb21c952e8989 |
||
|---|---|---|
| .. | ||
| files/ocf | ||
| lib/facter | ||
| manifests | ||
| spec | ||
| templates | ||
| .gitignore | ||
| .project | ||
| Gemfile | ||
| Modulefile | ||
| Rakefile | ||
| README | ||
This is a good start to play around with the galera multi-master mysql synchronous replication (http://www.codership.com/products/mysql_galera)
HOWTO:
* play around (add a database and data, chaos-monkey nodes etc.)
WARNING
Change the mysql root password in production and limit access to galera cluster members!
TODO
* naming: master -> donor
* put a load balancer in front of the cluster
* make cluster "masterless"
nodes.pp
# nodes
$cluster_name = 'wsrep_galera_cluster'
node /mysql-db-01/ {
class { 'galera' :
cluster_name => $cluster_name
}
}
node /mysql-db-0([2-9])/ {
$master_ip = '172.18.67.254'
class { 'galera' :
cluster_name => $cluster_name,
master_ip => $master_ip
}
}