error from gnt-cluster verify

[GoogleCodeExporter]

What software version are you running? Please provide the output of "gnt-
cluster --version", "gnt-cluster version", and "hspace --version".

What distribution are you using? CentOS 6.5 with KVM Installed Ganeti 
2.11(latest)

What steps will reproduce the problem?
1. gnt-cluster verify
2.
3.

What is the expected output? What do you see instead?

[root@node1 ganeti_webmgr]# gnt-cluster verify
Submitted jobs 17, 18
Waiting for job 17 ...
Sun May 11 10:25:18 2014 * Verifying cluster config
Sun May 11 10:25:18 2014 * Verifying cluster certificate files
Sun May 11 10:25:18 2014 * Verifying hypervisor parameters
Sun May 11 10:25:18 2014 * Verifying all nodes belong to an existing group
Waiting for job 18 ...
Sun May 11 10:25:18 2014 * Verifying group 'default'
Sun May 11 10:25:18 2014 * Gathering data (1 nodes)
Sun May 11 10:25:19 2014 * Gathering disk information (1 nodes)
Sun May 11 10:25:19 2014 * Verifying configuration file consistency
Sun May 11 10:25:19 2014 * Verifying node status
Sun May 11 10:25:19 2014   - ERROR: node node1.system.com: ssh communication 
with node 'node1.system.com': ssh problem: Permission denied 
(publickey,gssapi-keyex,gssapi-with-mic,password).\'r\n
Sun May 11 10:25:19 2014   - ERROR: node node1.system.com: the master node 
cannot reach the master IP (not configured?)
Sun May 11 10:25:19 2014 * Verifying instance status
Sun May 11 10:25:19 2014 * Verifying orphan volumes
Sun May 11 10:25:19 2014   - WARNING: node node1.system.com: volume 
vg_node1/lv_swap is unknown
Sun May 11 10:25:19 2014   - WARNING: node node1.system.com: volume 
vg_node1/lv_root is unknown
Sun May 11 10:25:19 2014 * Verifying N+1 Memory redundancy
Sun May 11 10:25:19 2014 * Other Notes
Sun May 11 10:25:19 2014 * Hooks Results

Please provide any additional information below.

Original issue reported on code.google.com by redhat.e...@gmail.com on 11 May 2014 at 2:26

[GoogleCodeExporter]

[root@node1 ~]# gnt-cluster getmaster
node1.system.com
[root@node1 ~]# ping node1.system.com
PING node1.system.com (192.168.1.200) 56(84) bytes of data.
64 bytes from node1.system.com (192.168.1.200): icmp_seq=1 ttl=64 time=0.029 ms
^C
--- node1.system.com ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 803ms
rtt min/avg/max/mdev = 0.029/0.029/0.029/0.000 ms

Original comment by redhat.e...@gmail.com on 11 May 2014 at 2:28

[GoogleCodeExporter]

[root@node1 ~]# gnt-node list
Node             DTotal DFree MTotal MNode MFree Pinst Sinst
node1.system.com 232.4G    0M   7.6G  1.0G  7.1G     0     0

Original comment by redhat.e...@gmail.com on 11 May 2014 at 2:28

[GoogleCodeExporter]

Concerning the  ssh-communication problem from node1.system.com to 
node1.system.com,
in order to debug it would be helpful to know how you created the cluster and
whether you changed something with the ssh-configuration afterwards (don't 
forget
automatic tools like puppet that also might modify ssh keys).

Concerning the master IP address, there is a difference between the master IP of
the cluster and IP of the master node. Essentially, the cluster name has to 
resolve
to an IP not used otherwise that is routed to the cluster. The reason for this
design is to have a canonical way to reach the cluster without knowing which
node currently is master.

Original comment by aeh...@google.com on 12 May 2014 at 8:18

• Changed state: NeedFeedback