Ceph入门到精通-大流量10GB/s支持OSPF(ECMP)-LVS 集群

Keepalived-LVS 能够提高集群的高可用性并增加后端检测功能、简化配置，满足常规需求。但Keepalived-LVS集群中，同一个VIP只能由一台设备进行宣告，为一主多备的架构，不能横向拓展集群的性能，为此我们引入OSPF来解决该问题。

OSPF(ECMP)

ECMP（Equal-CostMultipathRouting）等价多路径，存在多条不同链路到达同一目的地址的网络环境中，如果使用传统的路由技术，发往该目的地址的数据包只能利用其中的一条链路，其它链路处于备份状态或无效状态，并且在动态路由环境下相互的切换需要一定时间，而等值多路径路由协议可以在该网络环境下同时使用多条链路，不仅增加了传输带宽，并且可以无时延无丢包地备份失效链路的数据传输。

特点：

1. 基于流的均衡负载
2. 最大链路数受设备限制（最高16）
3. 所有链路都active，故障链路自动剔除

LVS+OSPF(ECMP)

利用ECMP以上特性，可以将LVS集群进行横向拓展，利用quagga启ospf
为模拟集群环境，我们准备了六台虚拟机分别为Client、LVS-1、LVS-2、RealServer1、RealServer2、Router，VIP设为192.168.0.100

• Router ：192.168.0.1 192.168.1.1
• Client ：192.168.1.2
• LVS-1 ：192.168.0.2
• LVS-2 ：192.168.0.3
• RealServer1 ：192.168.0.4
• RealServer2 ：192.168.0.5

Router 配置

LVS-1、LVS-2与Router需处于同一ospf域中，通过 IP 192.168.0.1 与LVS1/LVS2/RealServer1/RealServer2 通讯，IP 192.168.1.1 与 Client通讯

[root@router ~]# echo 1 >  /proc/sys/net/ipv4/ip_forward
// 开启转发
[root@router ~]# ifconfig
ens33: flags=4163  mtu 1500
        inet 192.168.0.1  netmask 255.255.255.0  broadcast 192.168.0.255
        inet6 fe80::20c:29ff:fe6e:d10e  prefixlen 64  scopeid 0x20
        ether 00:0c:29:6e:d1:0e  txqueuelen 1000  (Ethernet)
        RX packets 63921  bytes 5978914 (5.7 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 24354  bytes 2334494 (2.2 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ens38: flags=4163  mtu 1500
        inet 192.168.1.1  netmask 255.255.255.0  broadcast 192.168.1.255
        inet6 fe80::20c:29ff:fe6e:d122  prefixlen 64  scopeid 0x20
        ether 00:0c:29:6e:d1:22  txqueuelen 1000  (Ethernet)
        RX packets 60501  bytes 5206254 (4.9 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 18358  bytes 1432690 (1.3 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 1119  bytes 88568 (86.4 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1119  bytes 88568 (86.4 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
// ens33：192.168.0.1 与LVS1/LVS2/RealServer1/RealServer同一子网 ， ens38：192.168.1.1 与Client同一子网
[root@LVS1 ~]# yum install quagga.x86_64 -y
[root@router ~]# vi /etc/quagga/zebra.conf
hostname Router
password test
enable password test
log file /var/log/quagga/zebra.log
service password-encryption
interface ens33
interface ens38
access-list 1 permit 127.0.0.1
ip prefix-list ANY seq 5 permit 0.0.0.0/0 le 32
route-map ANY deny 10
  match ip address prefix-list ANY
ip protocol ospf route-map ANY
line vty
  access-class 1
[root@router ~]# vi /etc/quagga/ospfd.conf
hostname Router
password test
log file /var/log/quagga/ospfd.log
log stdout
log syslog
service password-encryption
interface ens33
  ip ospf hello-interval 1
  ip ospf dead-interval 4
  ip ospf priority 1
  ip ospf cost 1
router ospf
  ospf router-id 192.168.0.7
  log-adjacency-changes
  network 192.168.0.0/24 area 0.0.0.0
  network 192.168.1.0/24 area 0.0.0.0
access-list 1 permit 127.0.0.1
line vty
 access-class 1

RealServer 配置

realserver.sh 参考上一章节

[root@RealServer1 ~]# ./realserver.sh start
[root@RealServer1 ~]# ifconfig
ens33: flags=4163  mtu 1500
        inet 192.168.0.4  netmask 255.255.255.0  broadcast 192.168.0.255
        inet6 fe80::20c:29ff:febd:38da  prefixlen 64  scopeid 0x20
        ether 00:0c:29:bd:38:da  txqueuelen 1000  (Ethernet)
        RX packets 48635  bytes 4087456 (3.8 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 46101  bytes 5700308 (5.4 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 30  bytes 2613 (2.5 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 30  bytes 2613 (2.5 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo:0: flags=73  mtu 65536
        inet 192.168.0.100  netmask 255.255.255.255
        loop  txqueuelen 1000  (Local Loopback)

LVS 配置

由于我们使用ospf来实现高可用，不开启keepalived的vrrp功能（LVS2也同样配置），只是用其后端检测功能。
在keepalived-LVS集群的抢占模式下，Master在网卡上挂VIP并进行ARP广播，此时VIP对应的设备是唯一的。
但在OSPF-LVS集群中，Router根据ospf信息通过修改报文的目的mac地址转发到对应的LVS来实现均衡负载，并不根据VIP对应的ARP信息，所以对应的每台LVS将VIP挂在在lo上。

[root@LVS1 ~]# vi /etc/keepalived/keepalived.conf
global_defs {
        router_id LVS1 #路由器标识
        script_user root
        enable_script_security
}

virtual_server 192.168.0.100 80 {
        delay_loop 5
        lb_algo wrr
        lb_kind DR
        persistence_timeout 60
        persistence_granularity 255.255.255.255
        protocol tcp
        inhibit_on_failure on
        ha_suspend
        sorry_server 127.0.0.1 80
        real_server 192.168.0.4 80 {
            weight 10
                HTTP_GET{
            url{
                path /
                status_code 200
            }
        connect_port 80
            connect_timeout 2
            retry 1
            delay_before_retry 1
        }
        }
        real_server 192.168.0.5 80 {
            weight 10
                HTTP_GET{
            url{
                path /
                status_code 200
            }
        connect_port 80
            connect_timeout 2
            retry 1
            delay_before_retry 1
        }
        }
}
[root@LVS1 ~]# systemctl reload keepalived
[root@LVS1 ~]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  192.168.0.100:80 wrr persistent 60
  -> 192.168.0.4:80               Route   10     0          0
  -> 192.168.0.5:80               Route   10     0          0
[root@LVS1 ~]# ifconfig lo:0 192.168.0.100 netmask 255.255.255.255 up
//添加VIP到lo
[root@LVS1 ~]# ip a
1: lo:  mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet 192.168.0.100/32 scope global lo:0
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: ens33:  mtu 1500 qdisc pfifo_fast state UP group default qlen 1000
    link/ether 00:0c:29:af:6b:f7 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.2/24 brd 192.168.0.255 scope global noprefixroute ens33
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:feaf:6bf7/64 scope link
       valid_lft forever preferred_lft forever
[root@LVS1 ~]# route -n
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         192.168.0.1     0.0.0.0         UG    100    0        0 ens33
192.168.0.0     0.0.0.0         255.255.255.0   U     100    0        0 ens33

安装quagga并进行配置

[root@LVS1 ~]# yum install quagga.x86_64 -y
[root@LVS1 ~]# vi /etc/quagga/zebra.conf
hostname LVS1
# HOSTNAME改为IP也可以
password test
enable password test
log file /var/log/quagga/zebra.log
#log syslog
service password-encryption
interface ens33
access-list 1 permit 127.0.0.1
ip prefix-list ANY seq 5 permit 0.0.0.0/0 le 32
route-map ANY deny 10
  match ip address prefix-list ANY
ip protocol ospf route-map ANY
line vty
  access-class 1
[root@LVS1 ~]# vi /etc/quagga/ospfd.conf
hostname LVS1
password test
log file /var/log/quagga/ospfd.log
log stdout
log syslog
service password-encryption
interface ens33
  ip ospf hello-interval 1
  ip ospf dead-interval 4
  ip ospf priority 0
  ip ospf cost 1
router ospf
  ospf router-id 192.168.0.2
  log-adjacency-changes
  network 192.168.0.2/24 area 0.0.0.0
access-list 1 permit 127.0.0.1
line vty
 access-class 1
[root@LVS1 ~]# systemctl  start zebra
[root@LVS1 ~]# systemctl  start ospfd

同理我们配置完LVS2后进行检测ospf状态

[root@LVS1 ~]# vtysh
Hello, this is Quagga (version 0.99.22.4).
Copyright 1996-2005 Kunihiro Ishiguro, et al.
LVS1# show ip ospf  neighbor

    Neighbor ID Pri State           Dead Time Address         Interface            RXmtL RqstL DBsmL
192.168.0.7       1 Full/DR            3.817s 192.168.0.1     ens33:192.168.0.2        0     0     0
192.168.0.3       0 2-Way/DROther      3.518s 192.168.0.3     ens33:192.168.0.2        0     0     0
LVS1# show ip ospf  route
============ OSPF network routing table ============
N    192.168.0.0/24        [1] area: 0.0.0.0
                           directly attached to ens33
N    192.168.0.100/32      [1] area: 0.0.0.0
                           directly attached to lo
N    192.168.1.0/24        [2] area: 0.0.0.0
                           via 192.168.0.1, ens33

============ OSPF router routing table =============

============ OSPF external routing table ===========

在Route上查看路由

router# show ip route
Codes: K - kernel route, C - connected, S - static, R - RIP,
       O - OSPF, I - IS-IS, B - BGP, A - Babel,
       > - selected route, * - FIB route

K>* 0.0.0.0/0 via 192.168.0.10, ens33
C>* 127.0.0.0/8 is directly connected, lo
O   192.168.0.0/24 [110/1] is directly connected, ens33, 00:11:04
C>* 192.168.0.0/24 is directly connected, ens33
O   192.168.0.100/32 [110/2] via 192.168.0.2, ens33 inactive, 00:00:11
                             via 192.168.0.3, ens33 inactive, 00:00:11
O   192.168.1.0/24 [110/1] is directly connected, ens38, 00:12:19
C>* 192.168.1.0/24 is directly connected, ens38
//可以看到 到 192.168.0.100/32 的下一跳分配到 192.168.0.2、192.168.0.3

Client 测试

[root@Client ~]# route -n
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         192.168.1.1     0.0.0.0         UG    100    0        0 ens33
192.168.1.0     0.0.0.0         255.255.255.0   U     100    0        0 ens33
[root@Client ~]# traceroute 192.168.0.100
traceroute to 192.168.0.100 (192.168.0.100), 30 hops max, 60 byte packets
 1  192.168.1.1 (192.168.1.1)  0.575 ms  0.258 ms  0.478 ms
 2  192.168.0.100 (192.168.0.100)  1.901 ms  1.746 ms  1.370 ms
[root@Client ~]# curl 192.168.0.100
RealServer1 192.168.0.4

自动化

以上是最基础的ospf-Lvs集群配置，在实际生产环境中会有自动化部署、监控、告警等需求，例如quagga的初始化配置

quagga 配置

HOSTNAME=`hostname`
PASSWORD=shenyangchangkuan
#获取默认路由对应网卡
NIC=`/sbin/route -n|awk '$1=="default"||$1=="0.0.0.0"{print $NF}'|head -n 1`
if [ -z $NIC ];then
    echo "get NIC err,NIC is null!"
fi
#获取该网卡IP
#centos7改为下面这行
#IP=`/sbin/ifconfig $NIC |grep "inet "|sed 's/:/ /'|awk '{print $2}'`
IP=`/sbin/ifconfig $NIC|grep "inet addr"|sed 's/:/ /'|awk '{print $3}'`
if  [ -z $IP ];then
    echo "get IP err,IP is null!"
    exit 1
fi


#将该网卡及子网卡IP信息写入tmp/lvs_network.tmp
echo "">/tmp/lvs_network.tmp
for cfg in ` ls -l  /etc/sysconfig/network-scripts/ifcfg-${NIC}*|awk '{print $NF}'`
do
  cat $cfg |grep IPADDR=|awk -F "=" '{print $2}' |sed 's/"//g'>>/tmp/lvs_network.tmp
done