lxc_conf - conf.c:lxc_create_network:2888 - failed to create netdev

[Fathi]

Hi,

I have just upgraded one of my proxmoxve nodes from 4.4 to 5.0 whith the following command:
apt-get update && apt-get dist-upgrade -y && sed -i 's/jessie/stretch/g' /etc/apt/sources.list && sed -i 's/jessie/stretch/g' /etc/apt/sources.list.d/pve-enterprise.list && echo "deb http://download.proxmox.com/debian/ceph-luminous stretch main" > /etc/apt/sources.list.d/ceph.list ; apt-get update && apt-get dist-upgrade -y && reboot

After reboot, none of the contaniners/VM restarted and the node is greyed in the webgui of another node.
Now, when the outpout of pvecm nodes from the other node is:
Membership information
----------------------
Nodeid Votes Name
3 1 vortex
2 1 ovni7 (local)

and the outpout of pvecm status is:
Quorum information
------------------
Date: Tue Sep 26 18:45:38 2017
Quorum provider: corosync_votequorum
Nodes: 2
Node ID: 0x00000002
Ring ID: 3/6024
Quorate: Yes

Votequorum information
----------------------
Expected votes: 2
Highest expected: 2
Total votes: 2
Quorum: 2
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000003 1 10.1.10.101
0x00000002 1 10.1.12.111 (local)
So, my upgraded node seems to not be completely dead.

The same commands run from the upgraded node, gives the follwing output:
-bash: pvecm: command not found

Can someone tell me where my problems come from. I have already backed up the most important containers.

TIA

Fathi B.N.

 

[Alwin]

Can someone tell me where my problems come from.

Yes

apt-get update && apt-get dist-upgrade -y && sed -i 's/jessie/stretch/g' /etc/apt/sources.list && sed -i 's/jessie/stretch/g' /etc/apt/sources.list.d/pve-enterprise.list && echo "deb http://download.proxmox.com/debian/ceph-luminous stretch main" > /etc/apt/sources.list.d/ceph.list ; apt-get update && apt-get dist-upgrade -y && reboot

From this (especially '-y' tells apt "yes" to everything).

Check out and read carefully the wiki and docs before you do a upgrade.
https://pve.proxmox.com/wiki/Upgrade_from_4.x_to_5.0

 

[Fathi]

Yes


From this (especially '-y' tells apt "yes" to everything).

Check out and read carefully the wiki and docs before you do a upgrade.
https://pve.proxmox.com/wiki/Upgrade_from_4.x_to_5.0

Thank you Alwin, i read that document before doing the upgrade, just added '-y' to automate answers, as I did to upgrade from 4.3 to 4.4 and as I usually do.

 

[fabian]

Thank you Alwin, i read that document before doing the upgrade, just added '-y' to automate answers, as I did to upgrade from 4.3 to 4.4 and as I usually do.

if you tell apt to go ahead and do whatever it thinks is correct, then don't complain afterwards if apt does what it thinks is correct

you probably removed all/most of the PVE packages. I would suggest re-adding the repository configuration if it is missing, re-adding the repository key if it is missing, and then doing "apt update; apt install proxmox-ve"

 

[Fathi]

Thank you fabian, I did as you said and received a message about no candidate for package proxmox-ve, I updated sources.list to use pvetest and have a taste of 5.1, then did as you said. After some several reboot, I run an apt upgrade followed by another reboot. Now, I can access the webgui on the 5.1 beta node, but this node sees the 4.4 node as unavailable (greyed) and also the other node running 4.4 sees the upgraded 5.1beta node as unavailable (greyed).
pvecm expect 1 on both nodes did nothing returning Cannot initialize CMAP service, and I can't start nor delete any container as each node insists on "no quorum".
Also creating a new 5.0 instance and trying to join the cluster on any of the nodes outputs:
unable to copy ssh ID: cat: write error: Permission denied

My 2 nodes cluster is screwed up and I have to solve this issu before any of the containers running on the remaining cluster goes down.

 

[fabian]

sounds like your cluster is out of quorum or corosync is not running..