Search results

I have a question about volumes which VM that lives on NFS. Below is example: cluster1 vm101 vm102 cluster2 vm101 vm102 New Cluster - cluster3 - NO VM's on it yet. Now if I have All the VM's boots off SAME NFS volume that is attached to all 3 clusters. Is there a chance that the other...

yes, I fully understand what kickstart is, There another consideration that is FAI. https://fai-project.org/fai-cd/ It is also uses Debian.

This is resolved. Add memory was the fix. Thanks for telling how to look at the tasks as it was the critical piece I was missing.

Here is what found in tasks/0/directory: host10:/var/log/pve/tasks/0# more *qmstart:107* Total translation table size: 0 Total rockridge attributes bytes: 417 Total directory bytes: 0 Path table size(bytes): 10 Max brk space used 0 179 extents written (0 MB) kvm: cannot set up guest memory...

I see this in the active file listed in /var/log/pve/tasks directory: UPID:sc4-devops-qa-prxmx-host10:00005B74:01ED62BB:5E553D2F:qmstart:107:root@pam: 1 5E553D31 start failed: command '/usr/bin/kvm -id 107 -name workfource-ivasilyeu -chardev...

2020-02-25 19:34:31 starting migration of VM 107 to node 'host10' (x.x.x.21) 2020-02-25 19:34:31 copying disk images 2020-02-25 19:34:31 starting VM 107 on remote node 'host10' 2020-02-25 19:34:31 Debian GNU/Linux 9 2020-02-25 19:34:34 start failed: command '/usr/bin/kvm -id 107 -name ivasilyeu...

I do not see a task log at all. see below: host10:/etc/pve# ls -la total 13 drwxr-xr-x 2 root www-data 0 Dec 31 1969 . drwxr-xr-x 100 root root 195 Feb 24 02:16 .. -rw-r----- 1 root www-data 451 Oct 21 04:39 authkey.pub -r--r----- 1 root www-data 11999 Dec 31 1969 .clusterlog...

This issue is resolved. turns out the cluster name got changed. Thanks for responding.

Any updates

I forgot to post this: host07:~# qm config 107 | grep -v ssh boot: c bootdisk: scsi0 ciuser: root cores: 4 ide2: volume02:107/vm-107-cloudinit.qcow2,media=cdrom,size=4M ipconfig0: ip=x.x.x.209/22,gw=x.x.x..254 memory: 40960 name: workfource-ivasilyeu nameserver: x.x.x.15 net0...

Here is the source node: -host07:~# pveversion -v proxmox-ve: 5.4-2 (running kernel: 4.15.18-12-pve) pve-manager: 5.4-13 (running version: 5.4-13/aee6f0ec) pve-kernel-4.15: 5.4-12 pve-kernel-4.15.18-24-pve: 4.15.18-52 pve-kernel-4.15.18-12-pve: 4.15.18-36 corosync: 2.4.4-pve1 criu...

Will do, but the problem been fix. thanks

That did not work for me.

It turns out that corosync was messed up and had ensure corosync was corrected and make sure the version was right. keep in mind the /etc/pve directory is a virutal and part of the pmxcfs filesystem. It came back on it's own after I stop pve-cluster and corosync, then pmxcfs -l the restart...

You still miss my point totally why would I want to re-install and entire cluster? Really think about it. I understand what your saying just that it won't work, Real clusters such as Red Hat, VCS, Sun Cluster, HP Cluster do NOT have this issue. But they are not virtualized either. But neither...

Is there any kind of Migration checks that can be performed?

I got it working by cleaning up host10 and cleaning up corosync. To say the least this is most frustrating cluster I have EVER had the pleasure of dealing with.

I got this resolved. By cleaning up the node 10 I added and clean up the corosync. Corosync is far too sensitive.

i got it up finally after cleaning up the node I was adding.

I am trying to migration VM's back to my new node but not able to. Any suggestions: 2020-02-21 16:23:26 starting migration of VM 107 to node 'host10' (x.x.x..21) 2020-02-21 16:23:26 copying disk images 2020-02-21 16:23:26 starting VM 107 on remote node 'host10' 2020-02-21 16:23:30 start failed...