Linux 的iscsi网络存储以及配置Multipath 多路径存储

---

Linux 实现多路Multipath 多路径存储

环境说明：
应用服务端：proxy
ip规划：192.168.4.5/24 eth0
192.168.2.5/24 eth1
存储节点：client
ip规划： 192.168.4.10/24 eth0
192.168.2.5/24 eth1
multipath 概述 ：
当服务器到某一存储的设备有多条路径时u，每个路径就会识别为独立的设备
配置multipath就可以将服务器的节点和存储的阵列之间的多个I/O路径配置为单一设备，通过整合多个I/O路径，生成这些集合路径组成新的设备
以此达到冗余和负载均衡
前期配置
应用服务端：
[root@proxy ~]# rpm -q targetcli
targetcli-2.1.fb46-1.el7.noarch

创建ISCS存储：

	/> ls 
  [Storage Objects: 1]
  | | o- store ............... [/dev/vdb (20.0GiB) write-thru activated] [Targets: 1]
  | o- iqn.2019-03.cn.proxy:test ................................... [TPGs: 1]
  |   o- tpg1 .............................................. [no-gen-acls, no-auth]
  |     o- acls ............................................................ [ACLs: 1]
  |     | o- iqn.2019-03.cn.proxy:test ........... [Mapped LUNs: 1]
  |     |   o- mapped_lun0 ............................ [lun0 block/store(rw)]
  |     o- luns .................................................................. [LUNs: 1]
  |     | o- lun0 ..................................... [block/store (/dev/vdb)		(default_tg_pt_gp)]
  |     o- portals ......................................................... [Portals: 1]
  |       o- 0.0.0.0:3260 .............................................. [OK]
  o- loopback ........................................................... [Targets: 0]

  [root@proxy ~]# systemctl restart  target
   [root@proxy ~]# systemctl enable  target
  [root@proxy ~]# ss  -antpu  |grep  3260
			tcp    LISTEN     0      256       *:3260                  *:*     

存储节点

		[root@node ~]# rpm -q iscsi-initiator-utils.i686
							iscsi-initiator-utils.i686
      [root@node ~]# cat /etc/iscsi/initiatorname.iscsi  
					InitiatorName=iqn.2019-03.cn.proxy:test
	注意：节点的acl规则要跟应用服务端的acl设定一样。否则连接iscsi时会失败

( man iscsiadm )

 	[root@node ~]#iscsiadm --mode discoverydb --type sendtargets --portal 192.168.2.5 --discover
	[root@node ~]#iscsiadm --mode discoverydb --type sendtargets --portal 192.168.4.5 --discover
	

[root@node ~]# systemctl  restart  iscsi  
注意：有时会重启会报Warning！
	systemctl daemon-reload 后再重启即可
[root@node ~]# systemctl  restart  iscsid  
[root@node ~]# systemctl enable  iscsi  iscsid
[root@node ~]# sed -i  's/manual/automatic/'     /var/lib/iscsi/nodes/iqn.2019-03.cn.proxy:test/192.168.2.5\,3260\,1/default

`
[root@node ~]# rpm -q device-mapper-multipath
device-mapper-multipath-0.4.9-111.el7.x86_64

拷贝配置文件：
[root@node ~]# cp /usr/share/doc/device-mapper-multipath-0.4.9/multipath.conf /etc/

获取wwid：
[root@node ~]# /usr/lib/udev/scsi_id --whitelisted – \ device=/dev/sda
36001405f04a5a7a8d824b22855cc7f52

[root@node ~]# sed -n  '60,69p' /etc/multipath.conf 
multipaths {
	multipath {
       	wwid			36001405f04a5a7a8d824b22855cc7f52  //上一步获取的wwid
		alias			multi 	//multipath的自定义名字
		path_grouping_policy	multibus
		path_selector		"round-robin 0"
		failback		manual
		rr_weight		priorities
		no_path_retry		5
	}
	```
[root@node ~]# systemctl  restart multipathd.service 
*到这一步multipath的构建已经基本完成了！！！*

查看multipath：
[root@node ~]# multipath -l 
multi (36001405f04a5a7a8d824b22855cc7f52) dm-2 LIO-ORG ,store           
size=20G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=0 status=active
  |- 2:0:0:0 sda 8:0  active undef unknown
  `- 3:0:0:0 sdb 8:16 active undef unknown

接下来就可以当作是一块磁盘使用了，分区，格式化，挂载，使用即可

  ![在这里插入图片描述](https://img-blog.csdnimg.cn/2019030209594390.jpg?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80NDUwOTEzNA==,size_16,color_FFFFFF,t_70)