通过红帽的Piranha做集群，可是ipvsadm -ln没有realserver。

adminsinx

12

[ 本帖最后由 adminsinx 于 2009-3-20 15:10 编辑 ]

kns1024wh

pulse的日志放上来 cat /var/log/messages 中关于pulse的部分
/etc/sysconfig/ha/lvs.cfg 也放上来

adminsinx

lvs.cfg
----------------------------------------------------------------
primary = 10.10.21.10                     
#primary_private = 10.10.21.10            
service = lvs                                
backup_active = 1                           
backup = 10.10.21.11                     
#backup_private = 10.10.21.11            
heartbeat = 1                                
heartbeat_port = 539                        
keepalive = 3                                
deadtime = 12                                
network = direct                             
debug_level = NONE                           
virtual HTTP {                              
        active = 1                           
        address = 10.10.21.30 eth0:1   
        vip_nmask = 255.255.255.255         
#       fwmark = 80                          
        port = 80                           
        persistent = 320                     
        send = "GET /.lvs.html\r\n"         
        expect = "OK"
        use_regex = 1                              
        load_monitor = none                  
        scheduler = wlc                     
        protocol = tcp                       
        timeout = 5                          
        reentry = 10                        
        quiesce_server = 0                  
        server one {                        
               address = 10.10.21.20      
               active = 1                    
               weight = 100                  
        }                                    
        server two {                        
               address = 10.10.21.21      
               active = 1                    
               weight = 100                  
        }                                    
}                                            


log 又出现过正常的，现在都不正常了。
--------------------------------------------------------------
Mar 19 16:37:24 lvs2 pulse[6409]: gratuitous lvs arps finished
Mar 19 16:37:25 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:37:25 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:37:37 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:37:37 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:37:49 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:37:49 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:01 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:38:01 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:13 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:38:13 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:25 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:38:25 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:37 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:38:37 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:49 lvs2 nanny[6406]: READ to 10.10.21.20:80 timed out
Mar 19 16:38:49 lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out
Mar 19 16:38:49 lvs2 pulse[6401]: Terminating due to signal 15
Mar 19 16:38:49 lvs2 lvs[6403]: shutting down due to signal 15
Mar 19 16:38:49 lvs2 lvs[6403]: shutting down virtual service lvs
Mar 19 16:38:49 lvs2 nanny[6406]: Terminating due to signal 15
Mar 19 16:38:49 lvs2 nanny[6407]: Terminating due to signal 15
Mar 19 16:38:49 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 16:42:32 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 16:42:32 lvs2 pulse[6472]: STARTING PULSE AS BACKUP
Mar 19 16:56:48 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 16:56:48 lvs2 pulse[6615]: STARTING PULSE AS BACKUP
Mar 19 17:02:44 lvs2 pulse[6615]: Terminating due to signal 15
Mar 19 17:02:44 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 17:02:44 lvs2 pulse[6681]: STARTING PULSE AS BACKUP
Mar 19 17:05:07 lvs2 pulse[6681]: Terminating due to signal 15
Mar 19 17:05:07 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 17:05:07 lvs2 pulse[6712]: STARTING PULSE AS BACKUP
Mar 19 17:06:42 lvs2 pulse[6712]: partner dead: activating lvs
Mar 19 17:06:42 lvs2 lvs[6726]: starting virtual service HTTP active: 80
Mar 19 17:06:42 lvs2 nanny[6729]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:06:42 lvs2 lvs[6726]: create_monitor for HTTP/one running as pid 6729
Mar 19 17:06:42 lvs2 lvs[6726]: create_monitor for HTTP/two running as pid 6730
Mar 19 17:06:42 lvs2 nanny[6730]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:06:42 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:06:42 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:06:42 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:06:44 lvs2 setroubleshoot: SELinux is preventing ifconfig (ifconfig_t) "read write" to socket (initrc_t). For complete SELinux messages. run sealert -l b1e8e7dd-c402-46a0-bace-40d62cb12ea4
Mar 19 17:06:47 lvs2 pulse[6732]: gratuitous lvs arps finished
Mar 19 17:08:45 lvs2 pulse[6712]: Terminating due to signal 15
Mar 19 17:08:45 lvs2 lvs[6726]: shutting down due to signal 15
Mar 19 17:08:45 lvs2 lvs[6726]: shutting down virtual service HTTP
Mar 19 17:08:45 lvs2 nanny[6729]: Terminating due to signal 15
Mar 19 17:08:45 lvs2 nanny[6730]: Terminating due to signal 15
Mar 19 17:08:45 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:08:46 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 17:08:46 lvs2 pulse[6789]: STARTING PULSE AS BACKUP
Mar 19 17:08:47 lvs2 pulse[6789]: primary inactive (link failure?): activating lvs
Mar 19 17:08:47 lvs2 lvs[6791]: starting virtual service HTTP active: 80
Mar 19 17:08:47 lvs2 nanny[6796]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:08:47 lvs2 lvs[6791]: create_monitor for HTTP/one running as pid 6796
Mar 19 17:08:47 lvs2 nanny[6797]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:08:47 lvs2 lvs[6791]: create_monitor for HTTP/two running as pid 6797
Mar 19 17:08:47 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:08:47 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:08:47 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:08:47 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:08:47 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:08:52 lvs2 pulse[6793]: gratuitous lvs arps finished
Mar 19 17:10:42 lvs2 yum-updatesd-helper: error getting update info: Cannot retrieve repository metadata (repomd.xml) for repository: base. Please verify its path and try again
Mar 19 17:28:49 lvs2 pulse[6789]: Terminating due to signal 15
Mar 19 17:28:49 lvs2 lvs[6791]: shutting down due to signal 15
Mar 19 17:28:49 lvs2 lvs[6791]: shutting down virtual service HTTP
Mar 19 17:28:49 lvs2 nanny[6796]: Terminating due to signal 15
Mar 19 17:28:49 lvs2 nanny[6797]: Terminating due to signal 15
Mar 19 17:28:49 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:41:40 lvs2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Mar 19 17:41:40 lvs2 pulse[6942]: STARTING PULSE AS BACKUP
Mar 19 17:41:40 lvs2 pulse[6942]: primary inactive (link failure?): activating lvs
Mar 19 17:41:40 lvs2 lvs[6944]: starting virtual service HTTP active: 80
Mar 19 17:41:40 lvs2 nanny[6947]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:41:40 lvs2 lvs[6944]: create_monitor for HTTP/one running as pid 6947
Mar 19 17:41:40 lvs2 nanny[6948]: starting LVS client monitor for 10.10.21.30:80
Mar 19 17:41:40 lvs2 lvs[6944]: create_monitor for HTTP/two running as pid 6948
Mar 19 17:41:40 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:41:40 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:41:40 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:41:40 lvs2 avahi-daemon[5186]: Withdrawing address record for 10.10.21.30 on eth0.
Mar 19 17:41:40 lvs2 avahi-daemon[5186]: Registering new address record for 10.10.21.30 on eth0.
Mar 19 17:41:45 lvs2 pulse[6950]: gratuitous lvs arps finished

[ 本帖最后由 adminsinx 于 2009-3-19 17:54 编辑 ]

kns1024wh

配置文件中的

  send = "GET /.lvs.html\r\n"     

首先要明晰一点就是访问访问的url的确认信息是否正确被验证；
是否明确的使用了主从的配置，如果是就将主从上的日志都发上来分析；
是否已经执行

1. run sealert -l b1e8e7dd-c402-46a0-bace-40d62cb12ea4

复制代码

关键日志点:

lvs2 nanny[6407]: READ to 10.10.21.21:80 timed out

Mar 19 17:06:42 lvs2 pulse[6712]: partner dead: activating lvs  是在作切换测试么？

建议首先用pulse服务配置一台调度服务器，测试没有问题在配置主从的双机的

todayhero

nanny提示超时.表示你没有你的监控程序有问题.没有监牢到真实的服务器.所以将LVS中的清空

adminsinx

原帖由 kns1024wh 于 2009-3-19 21:07 发表
配置文件中的

首先要明晰一点就是访问访问的url的确认信息是否正确被验证；
是否明确的使用了主从的配置，如果是就将主从上的日志都发上来分析；
是否已经执行run sealert -l b1e8e7dd-c402-46a0-bace-40d ...

如何知道  ‘url的确认信息是否正确被验证’？

adminsinx

我查了下web的log。
[Fri Mar 20 09:02:46 2009] [error] [client 10.10.21.10] File does not exist: /var/www/html/.lvs.html

好像是没有通过验证。是不是哪里还需要设置？

kns1024wh

原帖由 adminsinx 于 2009-3-20 09:07 发表
我查了下web的log。
[Fri Mar 20 09:02:46 2009] [error] [client 10.10.21.10] File does not exist: /var/www/html/.lvs.html

好像是没有通过验证。是不是哪里还需要设置？

步骤这样把验证那个部分使用piranha默认的方式就可以了

adminsinx

改回默认的了。
现在的lvs.cfg
------------------------------------------
serial_no = 33
primary = 10.10.21.10
service = lvs
backup_active = 1
backup = 10.10.21.11
heartbeat = 1
heartbeat_port = 539
keepalive = 3
deadtime = 12
network = direct
debug_level = NONE
virtual HTTP {
     active = 1
     address = 10.10.21.30 eth0:1
     vip_nmask = 255.255.255.255
     port = 80
     persistent = 320
     send = "GET / HTTP/1.0rnrn"
     expect = "HTTP"
     use_regex = 1
     load_monitor = none
     scheduler = wlc
     protocol = tcp
     timeout = 5
     reentry = 10
     quiesce_server = 0
     server one {
         address = 10.10.21.20
         active = 1
         weight = 100
     }
     server two {
         address = 10.10.21.21
         active = 1
         weight = 100
     }
}


log
-----------------------------------------
Mar 20 09:54:02 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:02 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:54:12 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:12 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:54:22 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:22 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:54:32 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:32 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:54:42 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:42 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:54:52 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:54:52 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:02 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:02 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:12 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:12 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:22 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:22 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:32 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:32 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:42 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:42 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:55:52 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:55:52 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:02 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:02 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:12 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:12 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:22 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:22 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:32 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:32 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:42 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:42 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:56:52 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:56:52 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:02 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:02 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:12 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:12 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:22 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:22 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:32 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:32 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:42 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:42 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:57:52 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out
Mar 20 09:57:52 lvs1 nanny[6792]: READ to 10.10.21.21:80 timed out
Mar 20 09:58:02 lvs1 nanny[6791]: READ to 10.10.21.20:80 timed out

adminsinx

终于可以了。
原来我通过web设置会自动变成
send = "GET / HTTP/1.0rnrn"
改成  send = "GET / HTTP/1.0\r\n\r\n"
就好了。谢谢两位