ovirt-engine HA(active backup) cluster 구성 3 부

 

luci를 이용한 cluster share resouce 생성

luci webpage에서 …

Manage Clusters > Resouces > Add > IP Address

### !!! file system ID 는 비워둔다. (자동으로 들어감) ###

### file system resource는 위의 lvm 과 같게 만들어준다. 총 6개 !! ###

 

Manage Clusters > rhevm-cluster > service Groups > Add

Cluster 의 Service Group 생성 후 Add Resouce 를 클릭

192.168.144.30/24

ovirt ha lvm

etc-ovirt-engine

usr-share-ovirt-engine

usr-share-ovirt-engine

usr-share-ovirt-engine-wildfly

var-lib-ovirt-engine

var-lib-pgsql

postgresql

ovirt-engine-service

apache service

### 위와 같은 순서로 resouce 등록 (apache service 가 IP Address 보다 앞에 있으면 안됨)###

### /etc/cluster/cluster.conf (예제) ###

<?xml version=”1.0″?>

<cluster config_version=”33″ name=”rhevm-cluster”>

<clusternodes>

<clusternode name=”node1.test.dom” nodeid=”1″>

<fence>

<method name=”xvm fence”>

<device delay=”1″ domain=”225.0.1.12″ name=”kvm_xvm”/>

</method>

</fence>

</clusternode>

<clusternode name=”node2.test.dom” nodeid=”2″>

<fence>

<method name=”xvm fence”>

<device delay=”2″ domain=”225.0.1.12″ name=”kvm_xvm”/>

</method>

</fence>

</clusternode>

</clusternodes>

<cman expected_votes=”1″ two_node=”1″/>

<fencedevices>

<fencedevice agent=”fence_xvm” name=”kvm_xvm” timeout=”2″/>

</fencedevices>

<rm>

<failoverdomains>

<failoverdomain name=”ovirt_failover_domain” ordered=”1″ restricted=”1″>

<failoverdomainnode name=”node1.test.dom” priority=”1″/>

<failoverdomainnode name=”node2.test.dom” priority=”2″/>

</failoverdomain>

</failoverdomains>

<resources>

<ip address=”192.168.144.30/24″ sleeptime=”10″/>

<lvm name=”ovirt ha lvm” vg_name=”RHEVM”/>

<fs device=”/dev/RHEVM/etc-ovirt-engine” fsid=”63050″ fstype=”ext4″ mountpoint=”/etc/ovirt-engine” name=”etc-ovirt-engine” self_fence=”1″/>

<fs device=”/dev/RHEVM/usr-share-ovirt-engine” fsid=”45498″ fstype=”ext4″ mountpoint=”/usr/share/ovirt-engine” name=”usr-share-ovirt-engine” self_fence=”1″/>

<fs device=”/dev/RHEVM/usr-share-ovirt-engine-wildfly” fsid=”27022″ fstype=”ext4″ mountpoint=”/usr/share/ovirt-engine-wildfly” name=”usr-share-ovirt-engine-wildfly” self_fence=”1″/>

<fs device=”/dev/RHEVM/var-lib-ovirt-engine” fsid=”38611″ fstype=”ext4″ mountpoint=”/var/lib/ovirt-engine” name=”var-lib-ovirt-engine” self_fence=”1″/>

<fs device=”/dev/RHEVM/var-lib-pgsql” fsid=”47186″ fstype=”ext4″ mountpoint=”/var/lib/pgsql” name=”var-lib-pgsql” self_fence=”1″/>

<script file=”/etc/init.d/postgresql” name=”postgresql”/>

<script file=”/etc/init.d/ovirt-engine” name=”ovirt-engine-service”/>

<apache config_file=”conf/httpd.conf” name=”apache service” server_root=”/etc/httpd” shutdown_wait=”5″/>

<fs device=”/dev/RHEVM/etc-pki-ovirt-engine” fsid=”8507″ fstype=”ext4″ mountpoint=”/etc/pki/ovirt-engine” name=”etc-pki-ovirt-engine” self_fence=”1″/>

</resources>

<service domain=”ovirt_failover_domain” name=”ovirt-ha-cluster” recovery=”relocate”>

<ip ref=”192.168.144.30/24″/>

<lvm ref=”ovirt ha lvm”/>

<fs ref=”etc-ovirt-engine”/>

<fs ref=”etc-pki-ovirt-engine”/>

<fs ref=”usr-share-ovirt-engine”/>

<fs ref=”usr-share-ovirt-engine-wildfly”/>

<fs ref=”var-lib-ovirt-engine”/>

<fs ref=”var-lib-pgsql”/>

<script ref=”postgresql”/>

<script ref=”ovirt-engine-service”/>

<apache ref=”apache service”/>

</service>

</rm>

</cluster>

[root@node1 ~]# clusvcadm –r ovirt-ha-cluster

Trying to relocate service:ovirt-ha-cluster…Success

[root@node1 ~]# clustat

Cluster Status for rhevm-cluster @ Wed May 18 15:54:49 2016

Member Status: Quorate

Member Name ID Status

—— —- —- ——

node1.test.dom 1 Online, Local, rgmanager

node2.test.dom 2 Online, rgmanager

Service Name Owner (Last) State

——- —- —– —— —–

service:ovirt-ha-cluster node2.test.dom started

### cluster 절차 test. 이와 같이 나오면 성공!! ###

 

^ㅡ^