Difference between revisions of "ProxMox Manual Migration"
(Created page with "[http://www.google.com google is the best]") |
m |
||
| (One intermediate revision by the same user not shown) | |||
| Line 1: | Line 1: | ||
| − | [ | + | This will cover <strong>manually migrating config files for shared storage VM's</strong> from a dead [https://www.proxmox.com/en/proxmox-ve ProxMox] node to the remaining active one. This assumes that you have only two nodes and one has been disabled. In my particular case the SSD controller went out which crippled the node until I could get a new disk in. This is exactly why I made sure to use shared storage in my home lab. |
| + | |||
| + | The first thing you will need to do is set the quorum to expect only one vote (since you only have one node now). So issue the following command from the remaining live node | ||
| + | |||
| + | <code>pvecm expected 1</code> | ||
| + | |||
| + | Now you will need to move the configuration file to proper directory for it to become accessible on the remaining node. This command will be a variation depending on you specific version of [https://www.proxmox.com/en/proxmox-ve ProxMox]. I issued the "LS" (list) command several times to find the correct location of the configuration files I was looking for. In any case, it will be some variation of the following | ||
| + | |||
| + | <code>mv /etc/pve/nodes/node1/qemu-server/100.conf /etc/pve/nodes/node2/qemu-server</code> | ||
| + | |||
| + | Once the config file has been moved to the remaining node, you should see it switch over in the web gui. Issue the command for all of the configs until the directory is empty. If you have built you nodes correctly then one node should just be able to handle all critical VM's that are required to run. | ||
| + | |||
| + | At this point if you will be required to re-install the dead node from scratch then you will need to remove it from the cluster. Issue the following command to do so | ||
| + | |||
| + | <code>pvecm delnode node1</code> | ||
| + | |||
| + | This information was gathered from [http://pve.proxmox.com/pipermail/pve-user/2013-May/005616.html Here], [http://forum.proxmox.com/threads/8017-Migrate-CT-from-dead-cluster-node?p=45596#post45596 Here], & [https://pve.proxmox.com/wiki/Proxmox_VE_2.0_Cluster Here] | ||
| + | |||
| + | [[Category:ProxMox]] | ||
Latest revision as of 09:33, 24 November 2015
This will cover manually migrating config files for shared storage VM's from a dead ProxMox node to the remaining active one. This assumes that you have only two nodes and one has been disabled. In my particular case the SSD controller went out which crippled the node until I could get a new disk in. This is exactly why I made sure to use shared storage in my home lab.
The first thing you will need to do is set the quorum to expect only one vote (since you only have one node now). So issue the following command from the remaining live node
pvecm expected 1
Now you will need to move the configuration file to proper directory for it to become accessible on the remaining node. This command will be a variation depending on you specific version of ProxMox. I issued the "LS" (list) command several times to find the correct location of the configuration files I was looking for. In any case, it will be some variation of the following
mv /etc/pve/nodes/node1/qemu-server/100.conf /etc/pve/nodes/node2/qemu-server
Once the config file has been moved to the remaining node, you should see it switch over in the web gui. Issue the command for all of the configs until the directory is empty. If you have built you nodes correctly then one node should just be able to handle all critical VM's that are required to run.
At this point if you will be required to re-install the dead node from scratch then you will need to remove it from the cluster. Issue the following command to do so
pvecm delnode node1
