centos6.5 gfs2集群添加ctdb配置出错

q288288

集群搭建环境是Centos6.5+iscsi+cluster+gfs2+lvs+keepalived+ctdb(samba)
配置到ctdb就过不去了,google了ctdb配置，贴了日志文件请大家帮忙看看,winbind和samba相关服务自启动已经关掉,ip地址用到是公网ip

1. [root@physical2 Desktop]# ctdb status
   
2. Number of nodes:2
   
3. pnn:0 10.XX.XX.XX      UNHEALTHY
   
4. pnn:1 10.XX.XX.XX      UNHEALTHY (THIS NODE)
   
5. Generation:702696676
   
6. Size:2
   
7. hash:0 lmaster:0
   
8. hash:1 lmaster:1
   
9. Recovery mode:NORMAL (0)
   
10. Recovery master:1
    

复制代码

ctdb日志文件如下：

1. #vi /var/log/log.ctdb
   
2. 2014/12/03 03:30:24.340499 [57323]: Starting CTDBD (Version 2.5.1) as PID: 57323
   
3. 2014/12/03 03:30:24.570046 [57323]: Freeze priority 1
   
4. 2014/12/03 03:30:24.577522 [57323]: Freeze priority 2
   
5. 2014/12/03 03:30:24.577985 [57323]: Freeze priority 3
   
6. 2014/12/03 03:30:24.691923 [57323]: Freeze priority 1
   
7. 2014/12/03 03:30:24.692084 [57323]: Freeze priority 2
   
8. 2014/12/03 03:30:24.692139 [57323]: Freeze priority 3
   
9. 2014/12/03 03:30:28.504280 [recoverd:57458]: Taking out recovery lock from recovery daemon
   
10. 2014/12/03 03:30:28.504405 [recoverd:57458]: Take the recovery lock
    
11. 2014/12/03 03:30:28.517589 [57323]: Freeze priority 1
    
12. 2014/12/03 03:30:28.517712 [57323]: Freeze priority 2
    
13. 2014/12/03 03:30:28.517795 [57323]: Freeze priority 3
    
14. 2014/12/03 03:30:28.660214 [57323]: Thawing priority 1
    
15. 2014/12/03 03:30:28.660248 [57323]: Release freeze handler for prio 1
    
16. 2014/12/03 03:30:28.660262 [57323]: Thawing priority 2
    
17. 2014/12/03 03:30:28.660270 [57323]: Release freeze handler for prio 2
    
18. 2014/12/03 03:30:28.660280 [57323]: Thawing priority 3
    
19. 2014/12/03 03:30:28.660287 [57323]: Release freeze handler for prio 3
    
20. 2014/12/03 03:30:28.895209 [recoverd:57458]: Resetting ban count to 0 for all nodes
    
21. 2014/12/03 03:30:39.051798 [recoverd:57458]: server/ctdb_recoverd.c:3967 The vnnmap count is different from the number of active lmaster nodes: 1 vs 2
    
22. 2014/12/03 03:30:39.051846 [recoverd:57458]: Taking out recovery lock from recovery daemon
    
23. 2014/12/03 03:30:39.051856 [recoverd:57458]: Take the recovery lock
    
24. 2014/12/03 03:30:39.055349 [57323]: Freeze priority 1
    
25. 2014/12/03 03:30:39.056768 [57323]: Freeze priority 2
    
26. 2014/12/03 03:30:39.058577 [57323]: Freeze priority 3
    
27. 2014/12/03 03:30:39.184913 [57323]: Thawing priority 1
    
28. 2014/12/03 03:30:39.184954 [57323]: Release freeze handler for prio 1
    
29. 2014/12/03 03:30:39.184976 [57323]: Thawing priority 2
    
30. 2014/12/03 03:30:39.184986 [57323]: Release freeze handler for prio 2
    
31. 2014/12/03 03:30:39.184998 [57323]: Thawing priority 3
    
32. 2014/12/03 03:30:39.185006 [57323]: Release freeze handler for prio 3
    
33. 2014/12/03 03:30:39.440087 [recoverd:57458]: Resetting ban count to 0 for all nodes
    
34. 2014/12/03 03:30:53.461833 [recoverd:57458]: Trigger takeoverrun
    
35. 2014/12/03 03:30:54.383346 [57323]: 49.winbind: Starting Winbind services: /bin/bash: line 1: 57954 Aborted                 (core dumped) winbindd
    
36. 2014/12/03 03:30:54.390905 [57323]: 49.winbind: [FAILED]
    
37. 2014/12/03 03:30:54.391348 [57323]: 49.winbind: Failed to start winbind
    
38. 2014/12/03 03:30:54.391524 [57323]: startup event failed
    

复制代码

ctdb配置文件：

1. #vi /etc/sysconcifg/ctdb
   
2. CTDB_NODES=/etc/ctdb/nodes
   
3. CTDB_PUBLIC_ADDRESSES=/etc/ctdb/public_addresses
   
4. CTDB_RECOVERY_LOCK="/mnt/ctdb/.ctdb.lock"
   
5. CTDB_MANAGES_SAMBA=yes
   
6. CTDB_MANAGES_WINBIND=yes
   

复制代码