TwiX Renowned Member Feb 3, 2015 320 23 83 Aug 22, 2017 #1 Hello, I just want to know, on my pve 5 cluster, if is it possible to manually restart a non-ha vm (in case of failure node) ? Thanks in advanced, Antoine
Hello, I just want to know, on my pve 5 cluster, if is it possible to manually restart a non-ha vm (in case of failure node) ? Thanks in advanced, Antoine
wolfgang Proxmox Retired Staff Retired Staff Oct 1, 2014 6,496 563 103 Aug 23, 2017 #2 Hi, you mean on an other node? If so yes you can, but you have to copy the vm.config manually to the other node.
Hi, you mean on an other node? If so yes you can, but you have to copy the vm.config manually to the other node.
TwiX Renowned Member Feb 3, 2015 320 23 83 Aug 23, 2017 #3 Thanks, sounds good. This conf file is still readable when a node fail ? What happen when the node come back online ? Antoine
Thanks, sounds good. This conf file is still readable when a node fail ? What happen when the node come back online ? Antoine
wolfgang Proxmox Retired Staff Retired Staff Oct 1, 2014 6,496 563 103 Aug 23, 2017 #4 As long you have quorum yes. if you have no quorum you can set it quorate. see https://pve.proxmox.com/wiki/Cluster_Manager#_quorum
As long you have quorum yes. if you have no quorum you can set it quorate. see https://pve.proxmox.com/wiki/Cluster_Manager#_quorum
S spirit Distinguished Member Apr 2, 2010 7,035 1,132 273 www.groupe-cyllene.com Aug 29, 2017 #5 TwiX said: Thanks, sounds good. This conf file is still readable when a node fail ? What happen when the node come back online ? Antoine Click to expand... you have access of config of all vms, in all nodes, in /etc/pve/nodes/nodex/qemu-server/ simply do a move of the config like: - node2 crashed. on node 1 mv /etc/pve/nodes/node2/qemu-server/* /etc/pve/nodes/node1/qemu-server/ when node2 will restart, I'll sync /etc/pve with the current cluster config. (so vm will not be restarted on node2).
TwiX said: Thanks, sounds good. This conf file is still readable when a node fail ? What happen when the node come back online ? Antoine Click to expand... you have access of config of all vms, in all nodes, in /etc/pve/nodes/nodex/qemu-server/ simply do a move of the config like: - node2 crashed. on node 1 mv /etc/pve/nodes/node2/qemu-server/* /etc/pve/nodes/node1/qemu-server/ when node2 will restart, I'll sync /etc/pve with the current cluster config. (so vm will not be restarted on node2).