1

我在文件中有这些设置，keepalived.conf但是当我停止HAProxy服务时，它没有执行通知脚本，但是当我重新启动keepalived服务时，它每次都会执行。以下是详细信息，

HAProxy： 1.8.8

保活： 2.0.18

操作系统： Ubuntu 18.04

蟒蛇： 2.7

云服务提供商： Hetzner

/etc/keepalived/keepalived.conf

vrrp_script chk_haproxy {
    # Requires keepalived-1.1.13
    script "/usr/bin/pkill -0 haproxy"  # cheaper than pidof
    interval 2 # check every 2 seconds
    weight 2 # add 2 points of priority if OK
}

vrrp_instance real {
    interface eth0
    state MASTER
    virtual_router_id 51
    priority 101 # 101 on primary, 100 on secondary
    virtual_ipaddress {
        11.23.10.19/32 dev eth0 label eth0:1
    }
    track_script {
     chk_haproxy
    }

    notify "/etc/keepalived/master.sh"
    #notify_backup "/etc/keepalived/master.sh"
    #notify_fault "/etc/keepalived/master.sh"
}

/etc/keepalived/master.sh

#!/bin/bash

export API_TOKEN='<api_token>'
export MASTER_SERVER_ID='<master_server_id>'
export BACKUP_SERVER_ID='<backup_server_id>'
BASE_API='https://api.hetzner.cloud/v1'
FLOATING_IP_ID='<floating_ip_id>'
INSTANCE="Load-Balancer-Master"

if [ "$HOSTNAME" = "$INSTANCE" ]; then
    SERVER_ID=$BACKUP_SERVER_ID      # switch to the backup server if 
                                     # master gets down
else
    SERVER_ID=$MASTER_SERVER_ID      # vice-versa
fi
echo "Server ID: "  $SERVER_ID
HAS_FLOATING_IP=$(curl -H "Authorization: Bearer $API_TOKEN" -s 'https://api.hetzner.cloud/v1/servers/'$SERVER_ID|python -c "import sys,json; print( True if json.load(sys.stdin)['server']['public_net']['floating_ips'] else False)")
echo "Has Floating Ip: " $HAS_FLOATING_IP

if [ $HAS_FLOATING_IP = "False" ]; then
    n=0
    while [ $n -lt 10 ]
    do
        python /usr/local/bin/assign-ip $FLOATING_IP_ID $SERVER_ID && break
        n=$((n+1))
        sleep 3
    done
fi

/usr/local/bin/assign-ip

#!/usr/bin/python

import os
import sys
import requests
import json

api_base = 'https://api.hetzner.cloud/v1'


def usage():
    print('{0} [Floating IP] [Server ID]'.format(sys.argv[0]))
    print('\nYour Hetzner API token must be in the "API_TOKEN"'
          ' environmental variable.')


def main(floating_ip_id, server_id):
    payload = {'server': server_id}
    headers = {'Authorization': 'Bearer {0}'.format(os.environ['API_TOKEN']),
               'Content-type': 'application/json'}
    url = api_base + "/floating_ips/{0}/actions/assign".format(floating_ip_id)
    r = requests.post(url, headers=headers,  data=json.dumps(payload))

    resp = r.json()
    if resp['action']['error']:
        print('{0}: {1}'.format(resp['action']['command'], resp['error']['message']))
        sys.exit(1)
    else:
        print('Moving IP address to server: {0} with status:{1}'.format(server_id, resp['action']['status']))

if __name__ == "__main__":
    if 'API_TOKEN' not in os.environ or not len(sys.argv) > 2:
        usage()
        sys.exit()
    main(sys.argv[1], sys.argv[2])

当我停止使用 HAProxy 服务器sudo service haproxy stop并检查我收到此响应的状态时，

● haproxy.service - HAProxy Load Balancer
   Loaded: loaded (/lib/systemd/system/haproxy.service; enabled; vendor preset: enabled)
   Active: failed (Result: exit-code) since Sat 2019-09-28 22:12:57 IST; 1s ago
     Docs: man:haproxy(1)
           file:/usr/share/doc/haproxy/configuration.txt.gz
  Process: 26434 ExecStart=/usr/sbin/haproxy -Ws -f $CONFIG -p $PIDFILE $EXTRAOPTS (code=exited, stat
  Process: 26423 ExecStartPre=/usr/sbin/haproxy -f $CONFIG -c -q $EXTRAOPTS (code=exited, status=0/SU
 Main PID: 26434 (code=exited, status=143)

Sep 28 00:44:18 Load-Balancer-Master haproxy[26434]: Proxy nginx_pool started.
Sep 28 00:44:18 Load-Balancer-Master haproxy[26434]: Proxy nginx_pool started.
Sep 28 00:44:18 Load-Balancer-Master systemd[1]: Started HAProxy Load Balancer.
Sep 28 22:12:57 Load-Balancer-Master haproxy[26434]: [WARNING] 270/004418 (26434) : Exiting Master pr
Sep 28 22:12:57 Load-Balancer-Master haproxy[26434]: [ALERT] 270/004418 (26434) : Current worker 2643
Sep 28 22:12:57 Load-Balancer-Master haproxy[26434]: [WARNING] 270/004418 (26434) : All workers exite
Sep 28 22:12:57 Load-Balancer-Master systemd[1]: Stopping HAProxy Load Balancer...
Sep 28 22:12:57 Load-Balancer-Master systemd[1]: haproxy.service: Main process exited, code=exited, s
Sep 28 22:12:57 Load-Balancer-Master systemd[1]: haproxy.service: Failed with result 'exit-code'.
Sep 28 22:12:57 Load-Balancer-Master systemd[1]: Stopped HAProxy Load Balancer.

在/var/log/syslog我得到了这个，

Sep 28 18:35:41 Load-Balancer-Master systemd[1]: Started Session 114 of user driveu.
Sep 28 18:42:57 Load-Balancer-Master systemd[1]: Stopping HAProxy Load Balancer...
Sep 28 18:42:57 Load-Balancer-Master systemd[1]: haproxy.service: Main process exited, code=exited, status=143/n/a
Sep 28 18:42:57 Load-Balancer-Master systemd[1]: haproxy.service: Failed with result 'exit-code'.
Sep 28 18:42:57 Load-Balancer-Master systemd[1]: Stopped HAProxy Load Balancer.
Sep 28 18:42:57 Load-Balancer-Master Keepalived_vrrp[26884]: Script `chk_haproxy` now returning 1
Sep 28 18:42:57 Load-Balancer-Master Keepalived_vrrp[26884]: VRRP_Script(chk_haproxy) failed (exited with status 1)
Sep 28 18:42:57 Load-Balancer-Master Keepalived_vrrp[26884]: (real) Changing effective priority from 103 to 101

但是通知脚本不会被调用，浮动 ip 也不会分配给 BACKUP 实例。因为我真的很陌生，Keepalived任何人都可以帮我解决这个问题吗？

更新：我已经解决了这个问题

接口应该是private网络，并且必须指定MASTER和BACKUP使用unicast_src_ip和的私有ip unicast_peer。修改设置在这里，

vrrp_script chk_haproxy {
    # Requires keepalived-1.1.13
    script "/usr/bin/pkill -0 haproxy"  # cheaper than pidof
    interval 2 # check every 2 seconds
    weight 2 # add 2 points of priority if OK
}

vrrp_instance real {
    interface ens10      # changed it from eth0 
    state MASTER
    virtual_router_id 51
    priority 101 # 101 on primary, 100 on secondary
    unicast_src_ip 192.168.0.3
    unicast_peer {
        192.168.0.2
        }
    authentication {
            auth_type PASS
            auth_pass password
        }
    virtual_ipaddress {
        11.23.10.19/32 dev eth0 label eth0:1
    }
    track_script {
     chk_haproxy
    }

    notify "/etc/keepalived/master.sh"
    #notify_backup "/etc/keepalived/master.sh"
    #notify_fault "/etc/keepalived/master.sh"
}
4

0 回答 0