RHCA436-基于CentOS8pacemaker+corosync gfs2集群文件系统
配置集群文件系统
准备工作，指在红帽实验环境
lab start gfs-create
1.设置集群状态
pcs property set no-quorum-policy=freeze pcs property 查看配置 pcs property list --all pcs property show property man pacemaker-schedulerd 查看帮助 no-quorum-policy = enum [stop] What to do when the cluster does not have quorum What to do when the cluster does not have quorum Allowed values: stop, freeze, ignore, demote, suicide no-quorum-policy表示当达到小于最低仲裁数量时，默认集群会停止所有资源，freeze表示已经运行的resource还是运行的，但是不能加入新的resource了，集群保持原有状态，直到仲裁恢复；ignore表示忽略，继续运行
2.所有节点安装软件包
yum -y install gfs2-utils
3.格式化文件系统
mkfs.gfs2 -j 4 -t cluster1:gfsdata /dev/sharevg/sharelv1 -t：指定集群名称 -j：指定日志数量，不指定默认是1 -J：指定日志大小，最小和默认都是8M，最大1G，红帽推荐大于128M #查看日志大小 gfs2_edit -p journals /dev/mapper/sharevg-sharelv1 Block #Journal Status: of 524288 (0x80000) -------------------- Journal List -------------------- journal0: 0x12 8MB clean. journal1: 0x81b 8MB dirty. journal2: 0x1024 8MB clean. journal3: 0x182d 8MB clean. ------------------------------------------------------ #查看集群信息 [root@nodec ~]# tunegfs2 -l /dev/sharevg/sharelv1 tunegfs2 (Jun 9 2020 13:30:19) File system volume name: cluster1:gfsdata File system UUID: 9194a2ca-da3b-41bc-9611-a97a7db658e2 File system magic number: 0x1161970 Block size: 4096 Block shift: 12 Root inode: 9284 Master inode: 2072 Lock protocol: lock_dlm Lock table: cluster1:gfsdata
4.创建文件系统资源
pcs resource create clusterfs Filesystem device=/dev/sharevg/sharelv1 directory=/srv/data fstype=gfs2 op monitor interval=10s on-fail=fence --group=LVMshared
5.查看资源状态
pcs status --full
6.查看挂载情况
[root@nodea ~]# mount | grep gfs2 /dev/mapper/sharevg-sharelv1 on /srv/data type gfs2 (rw,relatime,seclabel) [root@nodeb ~]# mount | grep gfs2 /dev/mapper/sharevg-sharelv1 on /srv/data type gfs2 (rw,relatime,seclabel) [root@nodec ~]# mount | grep gfs2 /dev/mapper/sharevg-sharelv1 on /srv/data type gfs2 (rw,relatime,seclabel)
7.扩展文件系统
[root@nodea ~]# lvextend -L 2G /dev/sharevg/sharelv1 [root@nodea ~]# gfs2_grow -T /dev/sharevg/sharelv1 [root@nodea ~]# df -hT | grep gfs2 /dev/mapper/sharevg-sharelv1 gfs2 2.0G 37M 2.0G 2% /srv/data [root@nodeb ~]# df -hT | grep gfs2 /dev/mapper/sharevg-sharelv1 gfs2 2.0G 37M 2.0G 2% /srv/data [root@nodec ~]# df -hT | grep gfs2 /dev/mapper/sharevg-sharelv1 gfs2 2.0G 37M 2.0G 2% /srv/data
8.修改集群文件系统标签
tunegfs2 -o locktable=cluster1:gfs /dev/sharevg/sharelv1 tunegfs2 -l /dev/sharevg/sharelv1 | grep 'Lock table' Lock table: cluster1:gfsdata
配置安装httpd服务
在三个node中执行： yum install httpd -y firewall-cmd --add-service=http --permanent firewall-cmd --reload ssh root@nodea mkdir /var/www/html echo "RH436 Training" > /var/www/html/index.html restorcon -Rv /var/www/html
配置服务和ip资源
pcs resource create webip IPaddr2 ip=172.25.250.80 cidr_netmask=24 --group=myweb pcs resource create webserver apache --group=myweb
配置资源启动顺序
pcs constraint order start LVMshared-clone then myweb
