pvecm issue
- Thread starter CobraHost
- Start date
authentication key already exists
Get that when using the following.
pvecm add ****** -nodeid 2
Just uploaded a fix (pve-cluster_1.0-23_amd64.deb). Please can you update and test?
I/O warning : failed to load external entity "/etc/pve/cluster.conf"
ccs_tool: Error: unable to parse requested configuration file
command 'ccs_tool lsnode -c /etc/pve/cluster.conf' failed: exit code 1
unable to add node: command failed (ssh ******* -o BatchMode=yes pvecm addnode ***** --force 1 --nodeid 2)
when I remove /etc/pve/cluster.conf and then restart the cluster. it comes back the same says its a cluster any idea.
unable to open file '/etc/pve/nodes/*******/qemu-server/124.conf.tmp.92085' - Permission denied (500)
Also getting this on the same node as the one that is broken with the pvecm.
ccs_tool: Error: unable to parse requested configuration file
command 'ccs_tool lsnode -c /etc/pve/cluster.conf' failed: exit code 1
unable to add node: command failed (ssh ******* -o BatchMode=yes pvecm addnode ***** --force 1 --nodeid 2)
when I remove /etc/pve/cluster.conf and then restart the cluster. it comes back the same says its a cluster any idea.
unable to open file '/etc/pve/nodes/*******/qemu-server/124.conf.tmp.92085' - Permission denied (500)
Also getting this on the same node as the one that is broken with the pvecm.
Last edited:
You can't simply destroy a cluster by removing that file. There are other files involved.
Make sure the pve-cluster service is running (this is the filesystem mounted at /etc/pve):
# /etc/init.d/pve-cluster restart
Cannot reinstall the node as it's full of VPS that are running critical software
What? You are trying to add a node with existing VMs to the cluster (this will never work)?
Yes.
Just need to know how to remove the cluster so we can re-add because of the cluster not working in the first place, the whole system is messed up slightly. the only solution is backing up the VPS. how ever the main system does not even have enough space to do this.
Main Server we want the master to run on
root@*******:~# cat /etc/pve/cluster.conf
<?xml version="1.0"?>
<cluster name="KVMNodes" config_version="1">
<cman keyfile="/var/lib/pve-cluster/corosync.authkey">
</cman>
<clusternodes>
<clusternode name="*******" votes="1" nodeid="1"/>
</clusternodes>
</cluster>
2'd slave
cat /etc/pve/cluster.conf
cat: /etc/pve/cluster.conf: No such file or directory
Could I not send you the log in details of the server so that you could take a quick look, if not. I'm actually interested in purchasing the support if you can help fix these issues we are getting.
Last edited:
It is not possible to add a node with existing VMs to a cluster. You can only add 'empty' nodes.