GlusterFS 구축
시스템 정보
1 2 3 4 5 6 7 | OS : CentOS6.3 64bit vm : glusterFS1 : 192.168.0.184 glusterFS2 : 192.168.0.185 glusterFS-Client : 192.168.0.183 | cs |
구축 설정 정보
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | 서버노드 : 192.168.0.184~5 * 하나의 서버당 2대의 brick 으로 구성 (brick1/vol0 , brick1/vol1 brick1/vol2 ) -------------------------------------- pool : b brick : /brick1/vol0 - 폴더생성 /brick1/vol1 - 폴더생성 /brick1/vol2 - 폴더생성 volume name vol_dep volume type : replicated 2 client volume 마운트 : glusterfs | cs |
glusterFS (192.168.0.184) - 설정
time 동기화 (필수사항)
1 2 | rdate -s time.bora.net && /sbin/hwclock --systohc | cs |
glusterFS repo설정
1 2 3 4 5 | cd /etc/yum.repos.d/ wget http://download.gluster.org/pub/gluster/glusterfs/3.4/3.4.0/EPEL.repo/glusterfs-epel.repo | cs |
패키지 설치 & 서비스 시작
1 2 3 4 | yum install glusterfs-server glusterfs-geo-replication -y chkconfig --level 235 glusterd on service glusterd start | cs |
brick 디렉토리 생성
1 2 | mkdir -p /brick1/{vol0,vol1,vol2} -> 디렉토리로 생성하였음. | cs |
gluster pool 등록
1 2 3 | gluster peer probe 192.168.0.185 -> glusterFS1번을 제외한 나머지 서버노드 등록 | cs |
glusterFS 상태 확인
1 2 3 4 5 6 7 8 | [root@glusterfs1 vol1]# gluster peer status Number of Peers: 1 Hostname: 192.168.0.185 Uuid: 045ec7b0-241c-497b-9851-e2d75e49aa18 State: Peer in Cluster (Connected) | cs |
glusterFS volume 설정 - glusterfs1 번에서만 설정 (master)
1 2 3 | # gluster volume create vol_dep replica 2 transport tcp 192.168.0.184:/brick1/vol0/ 192.168.0.185:/brick1/vol0/ 192.168.0.184:/brick1/vol1/ 192.168.0.185:/brick1/vol1/ 192.168.0.184:/brick1/vol2/ 192.168.0.185:/brick1/vol2/ force -> volume name : vol_dep / type : replica 2 | cs |
부팅 후에도 마운트 될수 있게 rc.local에 등록
1 2 3 | # vi /etc/rc.local gluster volume create vol_dep replica 2 transport tcp 192.168.0.184:/brick1/vol0/ 192.168.0.185:/brick1/vol0/ 192.168.0.184:/brick1/vol1/ 192.168.0.185:/brick1/vol1/ 192.168.0.184:/brick1/vol2/ 192.168.0.185:/brick1/vol2/ force | cs |
Gluser volume 사용 시작
1 2 | #gluster volume start vol_dep | cs |
glusterFS volume 상태 확인
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | [root@glusterfs1 brick1]# gluster volume info all Volume Name: vol_dep Type: Distributed-Replicate Volume ID: 8a8d0b9d-5403-4bba-a479-509ff0e8b961 Status: Started Number of Bricks: 3 x 2 = 6 Transport-type: tcp Bricks: Brick1: 192.168.0.184:/brick1/vol0 Brick2: 192.168.0.185:/brick1/vol0 Brick3: 192.168.0.184:/brick1/vol1 Brick4: 192.168.0.185:/brick1/vol1 Brick5: 192.168.0.184:/brick1/vol2 Brick6: 192.168.0.185:/brick1/vol2 | cs |
glusterFS (192.168.0.185) - 설정
time 동기화 (필수사항)
1 2 | rdate -s time.bora.net && /sbin/hwclock --systohc | cs |
glusterFS repo설정
1 2 3 | cd /etc/yum.repos.d/ wget http://download.gluster.org/pub/gluster/glusterfs/3.4/3.4.0/EPEL.repo/glusterfs-epel.repo | cs |
패키지 설치 & 서비스 시작
1 2 3 4 | yum install glusterfs-server glusterfs-geo-replication -y chkconfig --level 235 glusterd on service glusterd start | cs |
brick 디렉토리 생성
1 2 | mkdir -p /brick1/{vol0,vol1,vol2} -> 디렉토리로 생성하였음. | cs |
gluster pool 등록
1 2 | gluster peer probe 192.168.0.184 -> glusterFS2번을 제외한 나머지 서버노드 등록 | cs |
glusterFS 상태 확인
1 2 3 4 5 6 7 8 | [root@glusterfs2 ~]# gluster peer status Number of Peers: 1 Hostname: 192.168.0.184 Uuid: 5a0c8ad9-8aae-418a-998c-b7298fa7fc6d State: Peer in Cluster (Connected) | cs |
glusterFS 상태 확인
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | [root@glusterfs2 brick1]# gluster volume info all Volume Name: vol_dep Type: Distributed-Replicate Volume ID: 8a8d0b9d-5403-4bba-a479-509ff0e8b961 Status: Started Number of Bricks: 3 x 2 = 6 Transport-type: tcp Bricks: Brick1: 192.168.0.184:/brick1/vol0 Brick2: 192.168.0.185:/brick1/vol0 Brick3: 192.168.0.184:/brick1/vol1 Brick4: 192.168.0.185:/brick1/vol1 Brick5: 192.168.0.184:/brick1/vol2 Brick6: 192.168.0.185:/brick1/vol2 | cs |
glusterFS (192.168.0.183) - 설정
time 동기화 (필수사항)
1 2 | rdate -s time.bora.net && /sbin/hwclock --systohc | cs |
패키지 설치
1 2 | #yum install glusterfs-fuse | cs |
volume 마운트
1 2 3 | # mkdir /mnt/gluster #mount -t glusterfs 192.168.0.184:vol_dep /mnt/gluster/ | cs |
mount 상태 확인
1 2 3 4 5 6 7 8 9 10 | [root@glusterfs-client gluster]# mount /dev/sda3 on / type ext4 (rw) proc on /proc type proc (rw) sysfs on /sys type sysfs (rw) devpts on /dev/pts type devpts (rw,gid=5,mode=620) tmpfs on /dev/shm type tmpfs (rw) /dev/sda1 on /boot type ext4 (rw) none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw) 192.168.0.184:vol_dep on /mnt/gluster type fuse.glusterfs (rw,default_permissions,allow_other,max_read=131072) | cs |
부팅시 마운드될수 있게 rc.local에 등록
1 2 3 | # vi /etc/rc.local mount -t glusterfs 192.168.0.184:vol_dep /mnt/gluster/ | cs |
기타
volume 삭제 방법
1 2 3 | #gluster volume stop <volume_name> #gluster volume delete <volume_name> | cs |
pool 추가 방법
1 | #gluster peer probe <pool_name> | cs |
volume 추가 방법
1 | #gluster volume add-brick <volume_name><add_server>:<add_directory> | cs |
volume 추가 후 용량 재 분배
1 2 | # gluster volume rebalance <vol_name> start # gluster volume rebalance <vol_name> status | cs |
brick 교체 방법
ex) server3:/exp3 의 brick을 server5:/exp3으로 변경
1 | #gluster volume replace-brick <volume_name> server3:/exp3 server5:/exp5 start | cs |
brick 제거 방법
1 | #gluster volume remove-brick <vol_name><server>:<directory> | cs |
성능튜닝
1 2 | #gluster volume set <volume_name><option><parameter> | cs |
* disk 성능측정은 iozone 으로 수행
