Bug 1403731

Description of problem:

Providing an invalid ip just before a valid ip in the volfile does not allow to do a mount,  neither cifs nor in windows.

glusterfs:volfile_server = 10.70.41.112 tcp+10.70.46.243:24007

Error messgae for a cifs mount : Resource temporarily unavailable
Error messgae for a Windows mount :  Specified network name is no longer available.

Where as with some ips provided as invalid ip, the mount succeeds. I dont know the difference between the 2 ips. eg : not working with 10.70.41.* and working with 10.70.47.*

This behavior seems a regression to previous builds. As this was not encountered with 3.8.4-6 and earlier.

**#Invalid IP : an ip that does not belong to the cluster and is not ping-able either
**#Valid IP : Ip of one of the server node that is ping-able from rest of nodes


I have attached two sets of logs.

Version-Release number of selected component (if applicable):
samba-client-libs-4.4.6-2.el7rhgs.x86_64
glusterfs-server-3.8.4-8.el7rhgs.x86_64
Windows 10


How reproducible:
Always

Steps to Reproduce:
1.On an existing 4 node setup with ctdb samba
2.Add the volfile server plugin with an invalid ip followed by a valid ip that belongs to one of the cluster nodes.
3.Try to do a cifs mount

Actual results:
cifs mount fails

Expected results:
cifs mount should succeed

Additional info:

Volume Name: hidenseek
Type: Distributed-Replicate
Volume ID: 0cb31b50-a742-49bd-bc1f-90eae7d911d5
Status: Started
Snapshot Count: 0
Number of Bricks: 2 x 2 = 4
Transport-type: tcp
Bricks:
Brick1: dhcp47-12.lab.eng.blr.redhat.com:/bricks/brick6/hidenseek_brick0
Brick2: dhcp47-120.lab.eng.blr.redhat.com:/bricks/brick6/hidenseek_brick1
Brick3: dhcp46-238.lab.eng.blr.redhat.com:/bricks/brick6/hidenseek_brick2
Brick4: dhcp46-243.lab.eng.blr.redhat.com:/bricks/brick6/hidenseek_brick3
Options Reconfigured:
performance.md-cache-timeout: 600
performance.cache-invalidation: on
performance.cache-samba-metadata: on
features.cache-invalidation-timeout: 600
features.cache-invalidation: on
storage.batch-fsync-delay-usec: 0
performance.stat-prefetch: on
server.allow-insecure: on
transport.address-family: inet
performance.readdir-ahead: on
nfs.disable: on
diagnostics.brick-log-level: DEBUG

*********************************************

Does not worked with
[gluster-hidenseek]
comment = For samba share of volume hidenseek
vfs objects = glusterfs
glusterfs:volume = hidenseek
glusterfs:logfile = /var/log/samba/glusterfs-hidenseek.%M.log
glusterfs:loglevel = 9
path = /
read only = no
guest ok = yes
glusterfs:volfile_server = 10.70.41.112 tcp+10.70.46.243:24007

Worked with
[gluster-hidenseek]
comment = For samba share of volume hidenseek
vfs objects = glusterfs
glusterfs:volume = hidenseek
glusterfs:logfile = /var/log/samba/glusterfs-hidenseek.%M.log
glusterfs:loglevel = 9
path = /
read only = no
guest ok = yes
glusterfs:volfile_server = 10.70.47.200 tcp+10.70.46.243:24007