Problem permission /etc/pve/nodes

[Jorge Fernando]

In my proxmox 3.4 installation I can not write to the /etc/pve directory.

root@proxmox26:/etc/pve# pveversion
pve-manager/3.4-1/3f2d890e (running kernel: 2.6.32-37-pve)

root@proxmox26:/etc/pve# pvecm status
cman_tool: Cannot open connection to cman, is it running ?

Please Help!!!

 

[t.lamprecht]

Note that I answered you there: https://forum.proxmox.com/threads/problem-permission-etc-pve-nodes.26999/#post-185203
But maybe lets keep thetopic here as the other thread is already solved an possibly another problem.

 

[Jorge Fernando]

Thanks for your reply, t.lamprecht.

Answering your question from:
https://forum.proxmox.com/threads/problem-permission-etc-pve-nodes.26999/#post-185203

root@proxmox26:/etc/pve# service cman restart
Stopping cluster:
   Stopping dlm_controld... [  OK  ]
   Stopping fenced... [  OK  ]
   Stopping cman... [  OK  ]
   Unloading kernel modules... [  OK  ]
   Unmounting configfs... [  OK  ]
Starting cluster:
   Checking if cluster has been disabled at boot... [  OK  ]
   Checking Network Manager... [  OK  ]
   Global setup... [  OK  ]
   Loading kernel modules... [  OK  ]
   Mounting configfs... [  OK  ]
   Starting cman... /usr/sbin/ccs_config_validate: line 186: 969541 Segmentation fault      (core dumped) ccs_config_dump > $tempfile

Unable to get the configuration
corosync [MAIN  ] Corosync Cluster Engine ('1.4.7'): started and ready to provide service.
corosync [MAIN  ] Corosync built-in features: nss
corosync [MAIN  ] Successfully read config from /etc/cluster/cluster.conf
corosync died with signal: 11 Check cluster logs for details
[FAILED]

root@proxmox26:/etc/pve# service pve-cluster restart
Restarting pve cluster filesystem: pve-cluster.

Another additional information is that I have a cluster with two nodes and from the web interface I only see the main in green. The other is shown in red, proxmox26 is the main node.

This happened a few weeks ago and I think it is related to loss of information due to electrical failures.

 

[t.lamprecht]

Starting cman... /usr/sbin/ccs_config_validate: line 186: 969541 Segmentation fault (core dumped) ccs_config_dump > $tempfile

Hmm, can you post your cluster configuration please? "/etc/pve/cluster.conf"

 

[Jorge Fernando]

/etc/pve/cluster.conf

<?xml version="1.0"?>
<cluster name="clusterHP" config_version="19">

  <cman keyfile="/var/lib/pve-cluster/corosync.authkey">
  </cman>

  <clusternodes>

  <clusternode name="proxmox27" votes="1" nodeid="2"/></clusternodes>

</cluster>

After an incorrect shutdown, due to a blackout the file /etc/pve/cluster.conf was unconfigured and I always modified it and the problem of the red node was solved, but now as I can not write to the /etc/pve directory I can do it.

I know that eventually I will reinstall that cluster, but I want to solve this problem if it has viable solution.

The main problem is I can not change /etc/pve/storage.cfg, and I need to add a new Storage.

 

[t.lamprecht]

<clusternodes> <clusternode name="proxmox27" votes="1" nodeid="2"/></clusternodes> </cluster>

OK, there is just one cluster node in there, which is a bit strange as you said you've got an two node cluster.

Try the cman config validator "ccs_config_validate" manually? I'm not really into the PVE 3.4 cluster stack, so I do not see directly the problem here.

Else dismantle the cluster for now and rebuild it:

service pve-cluster stop
pmxcfs -l      # <- starts the cluster file system in local mode where it should be readable
rm /etc/pve/cluster.conf # (there should be also such a file under /etc/cman or /etc/corosync or/var/lib/pve-cluster/ (sorry I pull this info from top of my head))
rm /var/lib/pve-cluster/corosync.authkey
killall pmxcfs
service pve-cluster start

Do this first on one node only, then when /etc/pve is OK after that and contains still all configs/... on the other node too.
You may then rebuild the cluster with `pvecm create clusterHP` on one and `pvecm add <node> --force` on the other.
Good luck I'm off for for a few days, so I hope someone else can help you if ther's still need.
Oh, and I really suggest going for 5.1 now, you get a bit more saner cluster stack and almost three years free of (big) update worries

 

[Jorge Fernando]

pvecm create clusterHP Error in main node (proxmox26)

root@proxmox26:/etc/pve# pvecm create clusterHP
Restarting pve cluster filesystem: pve-cluster[dcdb] crit: local cluster.conf is newer
.
Starting cluster:
   Checking if cluster has been disabled at boot... [  OK  ]
   Checking Network Manager... [  OK  ]
   Global setup... [  OK  ]
   Loading kernel modules... [  OK  ]
   Mounting configfs... [  OK  ]
   Starting cman... /usr/sbin/ccs_config_validate: line 186: 263606 Segmentation fault      ccs_config_dump > $tempfile

Unable to get the configuration
corosync [MAIN  ] Corosync Cluster Engine ('1.4.7'): started and ready to provide service.
corosync [MAIN  ] Corosync built-in features: nss
corosync [MAIN  ] Successfully read config from /etc/cluster/cluster.conf
corosync died with signal: 11 Check cluster logs for details
[FAILED]
root@proxmox26:/etc/pve#

2nd node proxmox27 , 10.10.88.10 is proxmox26

root@proxmox27:/mnt/pve/discoexterno# pvecm add --force 10.10.88.10
400 unable to parse boolean option
pvecm add <hostname> [OPTIONS]

without --force

root@proxmox27:/mnt/pve/discoexterno# pvecm add 10.10.88.10
this host already contains virtual machines - please remove the first

Help me please!!