Adding resources to the cluster

This is desirable as it can commit all changes in one atomic update.

Use the crm command prompt to do all configuration:

cib new sandbox

Sample Output:

node1:~# crm
crm(live)# cib new sandbox
INFO: sandbox shadow CIB created
crm(sandbox)#

This helps to decide if a resource should not run when the node has lost external network connectivity. It does that by providing a pingd attribute so that others can check on it to know the connection status.

configure primitive net_conn ocf:pacemaker:ping \
        params pidfile="/var/run/ping.pid" host_list="192.168.0.1" \
        op start interval="0" timeout="60s" \
        op stop interval="0" timeout="20s" \
        op monitor interval="10s" timeout="60s"

# run this resource on both node:
configure clone clone_net_conn net_conn \
        meta clone-node-max="1" clone-max="2"

One consistence IP address for the cluster regardless of which node is active or down.

configure primitive cluster_ip ocf:heartbeat:IPaddr2 \
  params ip="192.168.0.100" cidr_netmask="32" \
  op monitor interval="10s"

Don't run at the node which lost network connectivity: this rsc will stick with the drbd master, so no need the config below:

# disabled:
#configure location cluster_ip-no-conn cluster_ip \
#       rule $id="cluster_ip-no-conn-rule" -inf: not_defined pingd or pingd number:lte 0

This just run the crm_mon program in daemon mode. It provides:

• HTML status page
• E-mail notification of cluster events

configure primitive cluster_mon ocf:pacemaker:ClusterMon \
        params pidfile="/var/run/crm_mon.pid" \
        htmlfile="/var/tmp/crm_mon.html" \
        extra_options="-F ClusterMon@example.com --mail-to edp@example.com --mail-host smtp.example.com:25" \
        op start interval="0" timeout="20s" \
        op stop interval="0" timeout="20s" \
        op monitor interval="10s" timeout="20s"

The foundation layer of the VE: the DRBD device.

# we name the resource as "drbd_r0"
configure primitive drbd_r0 \
  ocf:linbit:drbd params drbd_resource="r0" \
  op start timeout="240s" \
  op stop timeout="100s" \
  op monitor role="Master" interval="11s" \
  op monitor interval="10s"

# we name the master/slave relation as "ms_drbd_r0"
configure ms ms_drbd_r0 drbd_r0 \
  meta master-max="1" master-node-max="1" \
  clone-max="2" clone-node-max="1" notify="true" \
  target-role="Started"

Don't run the master at the node which lost network connectivity:

configure location ms_drbd_r0-no-conn ms_drbd_r0 \
  rule $id="ms_drbd_r0-no-conn-rule" $role="Master" -inf: \
  not_defined pingd or pingd number:lte 0

Set node1 as the primary by preferring to run the DRBD Master on it:

configure location ms_drbd_r0-master-prefer-node1 ms_drbd_r0 \
  rule role="Master" 50: #uname eq node1

This filesystem is mounted on the /dev/drbd0 device provide by the drbd_r0 resource. The mount point is /vz.

configure primitive vz_fs ocf:heartbeat:Filesystem \
  params device="/dev/drbd0" \
  directory="/vz" \
  fstype="ext3" \
  op start timeout="60s" \
  op stop timeout="60s" \
  op monitor interval="10s" timeout="40s" 

configure primitive vz_svc lsb:vz \
  op monitor interval=10s 

Assume the ID of the created VE is 1011.

configure primitive ve1011 ocf:heartbeat:ManageVE \
  params veid=1011 \
  op start timeout=240s \
  op stop timeout=240s \  
  op monitor interval=10s 

Don't run at the node which lost network connectivity:

configure location ve1011-no-conn ve1011 \
        rule $id="ve1011-no-conn-rule" -inf: not_defined pingd or pingd number:lte 0

configure order ms_drbd_r0-b4-vz_fs inf: ms_drbd_r0:promote vz_fs:start
configure order vz_fs-b4-vz_svc inf: vz_fs:start vz_svc:start
configure order vz_svc-b4-ve1011 inf: vz_svc:start ve1011:start

Force these VE related resources to run together in the same node:

configure group group_vz vz_fs vz_svc ve1011
configure colocation coloc_vz inf: ms_drbd_r0:Master group_vz

Force the cluster management related resources to run together:

configure colocation coloc_mgmt inf: cluster_ip cluster_mon

Verify the configuration:

configure verify

If OK, commit the config to the live system:

cib commit sandbox

Execute at the shell:

crm << EOT
cib new sandbox

configure

primitive cluster_ip ocf:heartbeat:IPaddr2 \
        params ip="192.168.0.100" cidr_netmask="32" \
        op monitor interval="10s"
primitive cluster_mon ocf:heartbeat:ClusterMon \
        params pidfile="/var/run/crm_mon.pid" htmlfile="/var/tmp/crm_mon.html" \
        extra_options="-F ClusterMon@example.com --mail-to edp@example.com --mail-host smtp.example.com:25" \
        op start interval="0" timeout="20s" \
        op stop interval="0" timeout="20s" \
        op monitor interval="10s" timeout="20s"
primitive drbd_r0 ocf:linbit:drbd \
        params drbd_resource="r0" \
        op start interval="0" timeout="240s" \
        op stop interval="0" timeout="100s" \
        op monitor interval="11s" role="Master" \
        op monitor interval="10s"
primitive net_conn ocf:pacemaker:ping \
        params pidfile="/var/run/ping.pid" host_list="192.168.0.1" \
        op start interval="0" timeout="60s" \
        op stop interval="0" timeout="20s" \
        op monitor interval="10s" timeout="60s"
primitive ve1011 ocf:heartbeat:ManageVE \
        params veid="1011" \
        op start interval="0" timeout="240s" \
        op stop interval="0" timeout="240s" \
        op monitor interval="10s"
primitive vz_fs ocf:heartbeat:Filesystem \
        params device="/dev/drbd0" directory="/vz" fstype="ext3" \
        op start interval="0" timeout="60s" \
        op stop interval="0" timeout="60s" \
        op monitor interval="10s" timeout="40s"
primitive vz_svc lsb:vz \
        op monitor interval="10s"
group group_vz vz_fs vz_svc ve1011
ms ms_drbd_r0 drbd_r0 \
        meta master-max="1" master-node-max="1" clone-max="2" clone-node-max="1" notify="true" target-role="Started"
clone clone_net_conn net_conn \
        meta clone-node-max="1" clone-max="2"
location ms_drbd_r0-no-conn ms_drbd_r0 \
        rule $id="ms_drbd_r0-no-conn-rule" $role="Master" -inf: not_defined pingd or pingd number:lte 0 
location ms_drbd_r0-master-prefer-node1 ms_drbd_r0 \
        rule $id="ms_drbd_r0-master-prefer-node1-rule" $role="Master" 50: #uname eq node1
location ve1011-no-conn group_vz \
        rule $id="ve1011-no-conn-rule" -inf: not_defined pingd or pingd number:lte 0
colocation coloc_mgmt inf: cluster_ip cluster_mon
colocation coloc_vz inf: ms_drbd_r0:Master group_vz
order ms_drbd_r0-b4-vz_fs inf: ms_drbd_r0:promote group_vz:start

EOT

Verify and commit.

List installed resource agents:

crm ra list ocf heartbeat
crm ra list ocf pacemaker