Can't create new VM: no quroum

S

sam1994

New Member
Aug 17, 2014
17
1
1
HiMy VMs do not autostart and I cannot create a new VM (CT based on Proxmox 2.0). I once tried to set up a node but did not do so in the end. Here is my cluster.conf:
Code: 
root@vmhost:~# cat /etc/pve/cluster.conf
I try and run pvecm nodes to see what 'nodes' Proxmox thinks I have, and remove them. I get
Code: 
cman_tool: Cannot open connection to cman, is it running ?
so I try service cman start. Then I get:
Code: 
root@vmhost:~# service cman startStarting 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... Cannot find node name in cluster.confUnable to get the configurationCannot find node name in cluster.confcman_tool: corosync daemon didn't start Check cluster logs for details[FAILED]
My cluster.conf doesn't actually define any nodes, so I'm not sure how to proceed here. I just want to get Proxmox to allow me to create more CTs in the web interface, I do not want it to be part of a cluster.Thanks in advance
 
sam1994 said:
HiMy VMs do not autostart and I cannot create a new VM (CT based on Proxmox 2.0). I once tried to set up a node but did not do so in the end. Here is my cluster.conf:
Code: 
root@vmhost:~# cat /etc/pve/cluster.conf
I try and run pvecm nodes to see what 'nodes' Proxmox thinks I have, and remove them. I get
Code: 
cman_tool: Cannot open connection to cman, is it running ?
so I try service cman start. Then I get:
Code: 
root@vmhost:~# service cman startStarting 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... Cannot find node name in cluster.confUnable to get the configurationCannot find node name in cluster.confcman_tool: corosync daemon didn't start Check cluster logs for details[FAILED]
My cluster.conf doesn't actually define any nodes, so I'm not sure how to proceed here. I just want to get Proxmox to allow me to create more CTs in the web interface, I do not want it to be part of a cluster.Thanks in advance
Click to expand...
Hi,
/etc/pve/cluster.conf don't excist on an standalone pvehost!
Can you delete the empty file?

And that cman don't run is ok.

Udo
 
I can't delete cluster.conf, even as root!root@vmhost:~# rm /etc/pve/cluster.confrm: cannot remove `/etc/pve/cluster.conf': Permission deniedIt shouldn't exist, but I think I once tried to set it up as a cluster. Thanks
 
Ran those commandsnow I cannot even login in to the Proxmox environment.Connection error 401: Permission denied - invalid csrf token
 
Rebooted, now gettingAug 17 12:41:49 vmhost pmxcfs[2468]: [status] crit: cpg_send_message failed: 9in syslog constantly.pve-cluster service is already running (have checked this)this is bad. We are in a production environment!
 
This is my df outputudev 10M 0 10M 0% /devtmpfs 798M 332K 797M 1% /run/dev/mapper/pve-root 95G 1.2G 89G 2% /tmpfs 5.0M 0 5.0M 0% /run/locktmpfs 1.6G 3.1M 1.6G 1% /run/shm/dev/mapper/pve-data 341G 65G 277G 19% /var/lib/vz/dev/sda1 495M 37M 433M 8% /boot/dev/fuse 30M 12K 30M 1% /etc/pve
 
Hi,
is cman now running or not?
because if you are an one-node-cluster you can get quorum with "pvecm expected 1".

But this sounds that you don't do all steps of the linked post (backup /etc/pve; umount /etc/pve, remove cluster files; reboot).

Udo
 
Had to restore the backup of /etc/pve to get containers booting againweb interface now doesn't show containers at all
 
I have written a message on your profile. I understand if you don't take support messages, but would appreciate the help hereS
 
Hi,
I guess because your /etc/pve isn't mounted?!
Code: 
# on a single node
mount | grep /etc/pve
/dev/fuse on /etc/pve type fuse (rw,nosuid,nodev,relatime,user_id=0,group_id=0,default_permissions,allow_other)
ls -ls /var/lib/pve-cluster/
 84 -rw------- 1 root root  80896 Aug 13 22:17 config.db
 32 -rw------- 1 root root  32768 Aug 17 21:38 config.db-shm
140 -rw------- 1 root root 138368 Aug 17 21:38 config.db-wal

lsof | grep config.db
pmxcfs     2383            root  mem       REG              253,0     32768     164346 /var/lib/pve-cluster/config.db-shm
pmxcfs     2383            root    4u      REG              253,0     80896     164056 /var/lib/pve-cluster/config.db
...

/etc/init.d/pve-cluster restart
lsof | grep config.db
pmxcfs    17288             root  mem       REG              253,0     32768     164346 /var/lib/pve-cluster/config.db-shm
pmxcfs    17288             root    4u      REG              253,0     80896     164056 /var/lib/pve-cluster/config.db
...

# new process-id for the fuse-mounted /etc/pve
Udo
 
Udo has fixed the issue. Many thanks to him.Following his original link would have worked, however it should be noted when you mv /etc/pve to a backup directory, you should mkdir a fresh /etc/pve as restarting the service won't do this for you!
 
You must log in or register to reply here.
Top Bottom