midja****@tsuna*****
midja****@tsuna*****
2012年 1月 26日 (木) 15:43:43 JST
Fork松浦と申します。
初めて投稿させていただきます。
現在Heartbeat+DRBDにてクラスタ構成を検討しており、
下記のとおりフェイルオーバー動作する事を想定しています。
（１）node01/node02からルータ(192.168.0.1)に向けてPINGを送信し、
応答が無い場合はマスターノードのリソースをスレーブノードに切り替える。
（２）マスターノードのHeartbeatサービスを停止、またはノードをシャットダウンした際に、
リソースををスレーブノードに切り替える。
＜問題点＞
（１）はフェイルオーバーすることが確認出来ました。
（２）はスレーブノード上でDRBDリソースが一瞬マスターに切り替わるのですが、
すぐにスレーブに戻ってしまい、うまく切り替わりません。
＜環境＞
CentOS5.6 64bit
Pacemaker 1.0.11
Heartbeat 3.0.5
drbd 8.3.8
Pacemaker設定ファイル
-------------------------------------------------------------------------------------------
property $id="cib-bootstrap-options" \
cluster-infrastructure="openais" \
expected-quorum-votes="2" \
no-quorum-policy="ignore" \
stonith-enabled="false" \
startup-fencing="false" \
dc-deadtime="20s"
rsc_defaults $id="rsc-options" \
resource-stickiness="INFINITY" \
migration-threshold="1"
primitive res_drbd0 ocf:linbit:drbd \
params \
drbd_resource="r0" \
drbdconf="/etc/drbd.conf" \
op start interval="0s" timeout="240s" on-fail="restart" \
op monitor interval="11s" timeout="60s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" role="Master" \
op stop interval="0s" timeout="100s" on-fail="block"
primitive res_fs_drbd0 ocf:heartbeat:Filesystem \
params \
device="/dev/drbd0" \
directory="/chroot" \
fstype="ext3" \
op start interval="0s" timeout="60s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0s" timeout="60s" on-fail="block"
primitive res_vip ocf:heartbeat:IPaddr2 \
params \
nic="eth0" \
ip="192.168.0.111" \
cidr_netmask="24" \
op start interval="0s" timeout="90s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0s" timeout="100s" on-fail="block"
primitive res_ping ocf:pacemaker:pingd \
params \
name="default_ping_set" \
host_list="192.168.0.1" \
multiplier="100" \
dampen="0" \
meta \
migration-threshold="10" \
op start interval="0" timeout="90s" on-fail="restart" \
op monitor interval="10s" timeout="60s" on-fail="restart" \
op stop interval="0" timeout="100s" on-fail="ignore"
group rg_drbd \
res_vip res_fs_drbd0
ms ms_drbd0 res_drbd0 \
meta \
master-max="1" \
master-node-max="1" \
clone-max="2" \
clone-node-max="1" \
notify="true"
clone cl_ping res_ping \
meta \
clone-max="2" \
clone-node-max="1"
location loc_rg_drbd rg_drbd \
rule 200: #uname eq node01 \
rule 100: #uname eq node02 \
rule -INFINITY: defined default_ping_set and default_ping_set lt 100
location loc_ms_drbd0 ms_drbd0 \
rule 200: #uname eq node01 \
rule 100: #uname eq node02 \
rule role=master -INFINITY: defined default_ping_set and default_ping_set lt 100
colocation rg_on_drbd inf: rg_drbd ms_drbd0:Master
colocation cl_ping_col 1000: rg_drbd cl_ping
order ord_rg_aft_drbd inf: ms_drbd0:promote rg_drbd:start
-------------------------------------------------------------------------------------------
drbd.conf
-------------------------------------------------------------------------------------------
global { usage-count yes; }
common { syncer { rate 250M; } }
resource r0 {
protocol C;
startup {
degr-wfc-timeout 120;
}
net {
cram-hmac-alg sha1;
shared-secret "qawsedrftgyhujiko";
}
on node01 {
device /dev/drbd0;
disk /dev/sda8;
address 192.168.100.1:7789;
meta-disk internal;
}
on node02 {
device /dev/drbd0;
disk /dev/sda8;
address 192.168.100.2:7789;
meta-disk internal;
}
}
-------------------------------------------------------------------------------------------
Heartbeatサービスを停止した際の/var/log/messages
-------------------------------------------------------------------------------------------
Jan 26 14:49:38 node01 kernel: block drbd0: role( Primary -> Secondary )
Jan 26 14:49:43 node01 kernel: block drbd0: peer( Secondary -> Unknown ) conn( Connected -> Disconnecting ) pdsk( UpToDate -> DUnknown )
Jan 26 14:49:43 node01 kernel: block drbd0: short read expecting header on sock: r=-512
Jan 26 14:49:43 node01 kernel: block drbd0: meta connection shut down by peer.
Jan 26 14:49:43 node01 kernel: block drbd0: asender terminated
Jan 26 14:49:43 node01 kernel: block drbd0: Terminating asender thread
Jan 26 14:49:43 node01 kernel: block drbd0: Connection closed
Jan 26 14:49:43 node01 kernel: block drbd0: conn( Disconnecting -> StandAlone )
Jan 26 14:49:43 node01 kernel: block drbd0: receiver terminated
Jan 26 14:49:43 node01 kernel: block drbd0: Terminating receiver thread
Jan 26 14:49:43 node01 kernel: block drbd0: disk( UpToDate -> Diskless )
Jan 26 14:49:43 node01 kernel: block drbd0: drbd_bm_resize called with capacity == 0
Jan 26 14:49:43 node01 kernel: block drbd0: worker terminated
Jan 26 14:49:43 node01 kernel: block drbd0: Terminating worker thread
-------------------------------------------------------------------------------------------
Jan 26 14:49:38 node02 kernel: block drbd0: peer( Primary -> Secondary )
Jan 26 14:49:43 node02 kernel: block drbd0: peer( Secondary -> Unknown ) conn( Connected -> TearDown ) pdsk( UpToDate -> DUnknown )
Jan 26 14:49:43 node02 kernel: block drbd0: asender terminated
Jan 26 14:49:43 node02 kernel: block drbd0: Terminating asender thread
Jan 26 14:49:43 node02 kernel: block drbd0: Connection closed
Jan 26 14:49:43 node02 kernel: block drbd0: conn( TearDown -> Unconnected )
Jan 26 14:49:43 node02 kernel: block drbd0: receiver terminated
Jan 26 14:49:43 node02 kernel: block drbd0: Restarting receiver thread
Jan 26 14:49:43 node02 kernel: block drbd0: receiver (re)started
Jan 26 14:49:43 node02 kernel: block drbd0: conn( Unconnected -> WFConnection )
Jan 26 14:49:45 node02 kernel: block drbd0: role( Secondary -> Primary )
Jan 26 14:49:45 node02 kernel: block drbd0: Creating new current UUID
Jan 26 14:49:45 node02 kernel: block drbd0: State change failed: Need access to UpToDate data
Jan 26 14:49:45 node02 kernel: block drbd0: state = { cs:WFConnection ro:Primary/Unknown ds:UpToDate/DUnknown r--- }
Jan 26 14:49:45 node02 kernel: block drbd0: wanted = { cs:WFConnection ro:Primary/Unknown ds:Outdated/DUnknown r--- }
Jan 26 14:49:46 node02 kernel: block drbd0: role( Primary -> Secondary )
Jan 26 14:49:46 node02 kernel: block drbd0: disk( UpToDate -> Outdated )
-------------------------------------------------------------------------------------------
大変お手数ですが、どなたかご教授頂けますでしょうか。
以上よろしくおねがいします。