GBase8s CM配置

一、 单独CM

1. 在$GBASEDBTDIR/etc目录下创建sqlhosts.cm1文件，内容如下：

g_west    group   -       -       i=10

ol_54_pri onsoctcp 192.168.152.25 23701 g=g_west

ol_54_sds onsoctcp 192.168.152.26 23702 g=g_west

cm_update       group   -       -       i=12,c=0

oltp onsoctcp 192.168.152.25 25000 g=cm_update

2. 在$GBASEDBTDIR/etc目录下创建onconfig.cm1文件，内容如下：

NAME            cm1

LOGFILE         ${GBASEDBTDIR}/tmp/cm1.log

CM_TIMEOUT      40

SECONDARY_EVENT_TIMEOUT 20

EVENT_TIMEOUT   30

SQLHOSTS        LOCAL

LOG             1

DEBUG           1

CLUSTER main

{

 GBASEDBTSERVER        g_west

 SLA oltp    DBSERVERS=PRI+SDS   WORKERS=16

 FOC ORDER=ENABLED  TIMEOUT=10 RETRY=1  PRIORITY=1

}

3. 启停CM

启动cm

export  GBASEDBTSQLHOSTS=$GBASEDBTDIR/etc/sqlhosts.cm1

oncmsm -c $GBASEDBTDIR/etc/config.cm1

关闭CM

oncmsm -k cm1

二、 CM集群

1. 配置CM集群：

在主节点创建sqlhosts.cm1：内容

g_west  group   -       -       i=10

ol_54_pri onsoctcp 192.168.152.25 23701 g=g_west

ol_54_sds onsoctcp 192.168.152.26 23702 g=g_west

cm_update  group  -    -        i=12,c=0

oltp_all1       onsoctcp  192.168.152.25  23000 g=cm_update

oltp_all2       onsoctcp  192.168.152.26  23000 g=cm_update

cm_read   group  -    -        i=14,c=0

oltp_read1       onsoctcp  192.168.152.25  23001 g=cm_read

oltp_read2       onsoctcp  192.168.152.26  23001 g=cm_read

2. 创建etc/onconfig.cm1文件，内容如下：

NAME            cm1

LOGFILE         ${GBASEDBTDIR}/tmp/cm1.log

CM_TIMEOUT      40

SECONDARY_EVENT_TIMEOUT 20

EVENT_TIMEOUT   30

SQLHOSTS        LOCAL

LOG             1

DEBUG           1

CLUSTER main

{

 GBASEDBTSERVER         g_west

 SLA oltp_all1         DBSERVERS=PRI   WORKERS=16

 SLA oltp_read1        DBSERVERS=SDS   WORKERS=16

 FOC ORDER=ENABLED TIMEOUT=10 RETRY=1 PRIORITY=1

}

3. 更改 sqlhosts.cm1，onconfig.cm1两个文件的权限：

chown GBASEDBT:GBASEDBT sqlhosts.cm1 onconfig.cm1

chmod 660 sqlhosts.cm1 onconfig.cm1

4. 在命令行执行

export GBASEDBTDIR=/home/54_For_SSC/

export GBASEDBTSQLHOSTS=/home/54_For_SSC/etc/sqlhosts.cm1

5. 在SDS节点配置CM，创建etc/sqlhosts.cm2，内容如下：

g_west  group   -       -       i=10

ol_54_pri onsoctcp 192.168.152.25 23701 g=g_west

ol_54_sds onsoctcp 192.168.152.26 23702 g=g_west

cm_update  group  -    -        i=12,c=0

oltp_all1       onsoctcp  192.168.152.25  23000 g=cm_update

oltp_all2       onsoctcp  192.168.152.26  23000 g=cm_update

cm_read  group  -    -        i=14,c=0

oltp_read1       onsoctcp  192.168.152.25  23001 g=cm_read

oltp_read2       onsoctcp  192.168.152.26  23001 g=cm_read

6. 创建etc/config.cm2文件，内容如下：

NAME            cm2

LOGFILE         ${GBASEDBTDIR}/tmp/cm2.log

CM_TIMEOUT      40

SECONDARY_EVENT_TIMEOUT 20

EVENT_TIMEOUT   30

SQLHOSTS        LOCAL

LOG             1

DEBUG           1

CLUSTER main

{

 GBASEDBTSERVER         g_west

 SLA oltp_all2       DBSERVERS=PRI   WORKERS=16

 SLA oltp_read2      DBSERVERS=SDS   WORKERS=16

 FOC ORDER=ENABLED TIMEOUT=10 RETRY=1 PRIORITY=2}

7. 更改 sqlhosts.cm2，config.cm2两个文件的权限：

chown  GBASEDBT:GBASEDBT  sqlhosts.cm2  onconfig.cm2         

chmod 660 sqlhosts.cm2 onconfig.cm2

8. 在SDS节点执行下列语句

export GBASEDBTDIR=/home/54_For_SDS/

export GBASEDBTSQLHOSTS=/home/54_For_SDS/etc/sqlhosts.cm1

9. 在PRI节点启动cm1管理器：

oncmsm -c /home/54_For_SSC/etc/config.cm1

在SDS节点启动cm2管理器：

oncmsm -c /home/54_For_SSC/etc/config.cm2

10. 在集群中任意节点均可查看到cm状态。

onstat -g cmsm