samba+ctdb,Input/Output error

[shuai li]

Hi,guys

           I build samba+ctdb environment with 3 machines,the machine
192.168.4.79 acted as client,machine 192.168.4.88 and machine 192.168.4.237
acted as glusterfs and ctdb server,during the copy operation on client,I
shutdown a server to verify ctdb's virtual IP drifting.But after the
shutdown operation,Input/Output error occured on client.The detail process is
as follows:

First,on maching 192.168.4.88 and 192.168.4.237 ,I run command "ctdb
status" and "ctdb ip",follow information was showed:

[root at dcs88 data]# *ctdb status*
Number of nodes:2
pnn:0 192.168.4.237    OK
pnn:1 192.168.4.88     OK (THIS NODE)
Generation:1694962518
Size:2
hash:0 lmaster:0
hash:1 lmaster:1
Recovery mode:NORMAL (0)
Recovery master:0
[root at dcs88 data]# *ctdb ip*
Public IPs on node 1
192.168.4.109 1
192.168.4.110 0


[root at dcs237 data]#* ctdb status*
Number of nodes:2
pnn:0 192.168.4.237    OK (THIS NODE)
pnn:1 192.168.4.88     OK
Generation:1694962518
Size:2
hash:0 lmaster:0
hash:1 lmaster:1
Recovery mode:NORMAL (0)
Recovery master:0
[root at dcs237 data]# *ctdb ip*
Public IPs on node 0
192.168.4.109 node[1] active[] available[eth0] configured[eth0]
192.168.4.110 node[0] active[eth0] available[eth0] configured[eth0]

from the information we know that two servers was running normally,and the
virtual ip 192.168.4.109 was controled by machine 192.168.4.88 and virtual
ip 192.168.4.110 was controled by machine 192.168.4.237.


Second,on machine 192.168.4.79,I mounted 192.168.4.109 with cifs
protoacl.After typed mount command,follow information was showd,we can find
that 192.168.4.109 was mounted successfully.

[root at dcs79 mnt]# *mount *
/dev/mapper/vg_dcs2-lv_root on / type ext4 (rw)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw)
/dev/sda1 on /boot type ext4 (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)
none on /sys/kernel/config type configfs (rw)
nfsd on /proc/fs/nfsd type nfsd (rw)
gvfs-fuse-daemon on /root/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev)
/dev/sr0 on /media/CentOS_6.4_Final type iso9660
(ro,nosuid,nodev,uhelper=udisks,uid=0,gid=0,iocharset=utf8,mode=0400,dmode=0500)
*192.168.4.109:/public on /mnt/data type cifs (rw)*

Thirdly,on machine 192.168.4.79,I copyed a file to mount point,during the
process,I runned "service ctdb stop"command on 192.168.4.88,servel seconds
later,an error occurs:

[root at dcs79 data]# cp /root/test ./
cp: closing `./test': Input/output error

Last,on machine 192.168.4.237,after running "ctdb ip"command,follow
information was showed:
[root at dcs237 data]# *ctdb status*
Number of nodes:2
pnn:0 192.168.4.237    OK (THIS NODE)
pnn:1 192.168.4.88     DISCONNECTED|UNHEALTHY|INACTIVE
Generation:1637129627
Size:1
hash:0 lmaster:0
Recovery mode:NORMAL (0)
Recovery master:0
[root at dcs237 data]# *ctdb ip*
Public IPs on node 0
192.168.4.109 node[0] active[eth0] available[eth0] configured[eth0]
192.168.4.110 node[0] active[eth0] available[eth0] configured[eth0]

The virtual IP 192.168.4.109 was takeover by machine 192.168.4.237
successfully,but copy operation still killed by Input/Output error  on
client.

I hope someone can help me!

Best regards!!Thank you!!




shuai li