RHEL8 多路径配置详解(二)

前言

在企业级数据中心中，Fibre Channel (FC) 技术因其高速度、低延迟和高可靠性而广泛应用于存储网络。然而，单一的FC链路存在单点故障风险，一旦链路或相关硬件出现故障，可能导致数据访问中断，影响业务连续性。

为了解决这一问题，多路径和双链路策略成为构建高可用性FC存储网络的关键技术，本文将模拟FC双链路存储网络。

一、环境配置

操作系统	主机名	IP地址
rhel8	iscsi-server	192.168.2.115
		192.168.2.125
rhel8	iscsi-client	192.168.2.116

1.1 iscsi-server

• 网络配置

[root@db1 ~]# ip a

2: ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000

    link/ether 00:0c:29:1d:d3:ac brd ff:ff:ff:ff:ff:ff

    inet 192.168.2.115/24 brd 192.168.2.255 scope global noprefixroute ens160

      valid_lft forever preferred_lft forever

    inet6 fe80::890c:e836:972e:6de/64 scope link dadfailed tentative noprefixroute 

      valid_lft forever preferred_lft forever

    inet6 fe80::5a22:4f3f:10e4:92bb/64 scope link noprefixroute 

      valid_lft forever preferred_lft forever

3: ens224: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000

    link/ether 00:0c:29:1d:d3:b6 brd ff:ff:ff:ff:ff:ff

    inet 192.168.2.125/24 brd 192.168.2.255 scope global noprefixroute ens224

      valid_lft forever preferred_lft forever

    inet6 fe80::a48a:2770:ffb9:a2e2/64 scope link noprefixroute 

      valid_lft forever preferred_lft forever

• 存储配置

Disk /dev/sda: 2 GiB, 2147483648 bytes, 4194304 sectors

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

/dev/sda作为存储设备对外提供服务

1.2 iscsi-client网络配置

[root@db2 ~]# ip a

2: ens160: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000

    link/ether 00:0c:29:87:45:61 brd ff:ff:ff:ff:ff:ff

    inet 192.168.2.116/24 brd 192.168.2.255 scope global noprefixroute ens160

      valid_lft forever preferred_lft forever

    inet6 fe80::890c:e836:972e:6de/64 scope link noprefixroute 

      valid_lft forever preferred_lft forever

二、配置iscsi-server

由于我们没有存储设备，所以选择使用ISCSI块存储服务来模拟存储设备，通过在iscsi-server主机上绑定两个ip地址，来模拟存储设备到主机client的双链路。

2.1 关闭防火墙和selinux

--关闭防火墙

systemctl stop firewalld

systemctl disabled firewalld

--编辑/etc/selinux/config 

设置SELINUX=disabled

reboot

2.2 安装ISCSI块存储服务

[root@db1 ]# yum install -y targetcli

2.3 配置ISCSI块存储服务

[root@db1 network-scripts]# targetcli

Warning: Could not load preferences file /root/.targetcli/prefs.bin.

targetcli shell version 2.1.53

Copyright 2011-2013 by Datera, Inc and others.

For help on commands, type 'help'.

/> /backstores/block create dream:storage1 /dev/sda

Created block storage object dream:storage1 using /dev/sda.

/> /iscsi create iqn.2024-08.com.example:iscsi-server   

Created target iqn.2024-08.com.example:iscsi-server.

Created TPG 1.

Global pref auto_add_default_portal=true

Created default portal listening on all IPs (0.0.0.0), port 3260.

/iscsi> cd iqn.2024-08.com.example:iscsi-server/tpg1/acls 

/iscsi/iqn.20...ver/tpg1/acls> /iscsi/iqn.2024-08.com.example:iscsi-server/tpg1/acls create iqn.2024-08.com.example:iscsi-server

Created Node ACL for iqn.2024-08.com.example:iscsi-server

/iscsi/iqn.20...ver/tpg1/acls> ls

o- acls ...................................................................... [ACLs: 1]

  o- iqn.2024-08.com.example:iscsi-server ............................. [Mapped LUNs: 0]

/iscsi/iqn.20...ver/tpg1/acls> /iscsi/iqn.2024-08.com.example:iscsi-server/tpg1/luns create /backstores/block/dream:storage1     

Created LUN 0.

Created LUN 0->0 mapping in node ACL iqn.2024-08.com.example:iscsi-server

/iscsi/iqn.20...ver/tpg1/acls> ls

o- acls ...................................................................... [ACLs: 1]

  o- iqn.2024-08.com.example:iscsi-server ............................. [Mapped LUNs: 1]

    o- mapped_lun0 .................................... [lun0 block/dream:storage1 (rw)]

/iscsi/iqn.20...ver/tpg1/acls> exit

Global pref auto_save_on_exit=true

Configuration saved to /etc/target/saveconfig.json

2.4 启动服务

[root@db1 network-scripts]# systemctl start target

[root@db1 network-scripts]# systemctl status target

● target.service - Restore LIO kernel target configuration

  Loaded: loaded (/usr/lib/systemd/system/target.service; disabled; vendor preset: dis>

  Active: active (exited) since Wed 2024-07-31 10:36:30 CST; 3s ago

  Process: 9009 ExecStart=/usr/bin/targetctl restore (code=exited, status=0/SUCCESS)

 Main PID: 9009 (code=exited, status=0/SUCCESS

三、配置iscsi客户端

3.1 安装 iscsi-initiator-utils

[root@db2 ~]# yum install -y iscsi-initiator-utils

3.2 加入iscsi认证

[root@db2 ~]# vim /etc/iscsi/initiatorname.iscsi

InitiatorName=iqn.2024-08.com.example:iscsi-server

3.3 发现服务端ISCSI设备

[root@db2 ~]# iscsiadm -m discovery -t st -p 192.168.2.115

192.168.2.115:3260,1 iqn.2024-08.com.example:iscsi-server

[root@db2 ~]# iscsiadm -m discovery -t st -p 192.168.2.125

192.168.2.125:3260,1 iqn.2024-08.com.example:iscsi-server

3.4 登陆设备

[root@db2 ~]# iscsiadm -m node -T iqn.2024-08.com.example:iscsi-server -p 192.168.2.115 -l

Logging in to [iface: default, target: iqn.2024-08.com.example:iscsi-server, portal: 192.168.2.115,3260]

Login to [iface: default, target: iqn.2024-08.com.example:iscsi-server, portal: 192.168.2.115,3260] successful.

[root@db2 ~]# 

[root@db2 ~]# iscsiadm -m node -T iqn.2024-08.com.example:iscsi-server -p 192.168.2.125 -l

Logging in to [iface: default, target: iqn.2024-08.com.example:iscsi-server, portal: 192.168.2.125,3260]

Login to [iface: default, target: iqn.2024-08.com.example:iscsi-server, portal: 192.168.2.125,3260] successful.

3.5 查看磁盘信息

Disk /dev/sdd: 2 GiB, 2147483648 bytes, 4194304 sectors

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 4194304 bytes

Disk /dev/sde: 2 GiB, 2147483648 bytes, 4194304 sectors

Units: sectors of 1 * 512 = 512 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 4194304 bytes

本质上两个存储设备为同一个，但被识别成了两块磁盘

[root@db2 ~]# /usr/lib/udev/scsi_id --whitelisted --device=/dev/sdd

360014059ca320028d97460499522865b

[root@db2 ~]# /usr/lib/udev/scsi_id --whitelisted --device=/dev/sde

360014059ca320028d97460499522865b

四、配置多路径

安装在上篇已经介绍过，这里不再赘述。

4.1 多路径配置文件

[root@db2 ~]# cat /etc/multipath.conf

defaults {

    user_friendly_names yes

    find_multipaths yes

  }

  blacklist {

}

multipaths {

    multipath {

    wwid   360014059ca320028d97460499522865b

    alias  mpsda

}

4.2 重启生效

[root@db2 ~]# multipath -r

[root@db2 ~]# multipath -ll

mpsda (360014059ca320028d97460499522865b) dm-4 LIO-ORG,dream:storage1

size=2.0G features='0' hwhandler='1 alua' wp=rw

|-+- policy='service-time 0' prio=50 status=active

| `- 33:0:0:0 sdd     8:48  active ready running

`-+- policy='service-time 0' prio=50 status=enabled

  `- 34:0:0:0 sde     8:64  active ready running

结语

通过实施多路径和双链路策略，企业数据中心可以构建高可用性FC存储网络，确保数据的连续访问和高性能传输，从而支撑关键业务的连续运行。随着技术的不断发展，多路径和双链路策略将继续在存储网络中发挥重要作用，助力企业构建更加强健、灵活的IT基础设施。