[SOLVED] Error 500 cluster not ready. Please help

[thecharliewahl]

Hello, I had a cluster, one of the servers disappeared. After that I cant run any of my vms. And inside of the cluster tab the cluster information is grayed out but it still shows that there are two nodes. When I look at the status of Corosync it says "Could not open /etc
corosync/authkey: No such file or directory". I NEED to get the data off of one of the VMs that I use as a nas.

 

[Moayad]

Hi,

How did the server disappear? Did you remove it from the cluster?

Can you post the syslog since the issue happened? You can sort the syslog with a specific time/date using journalctl e.g.:

journalctl --since "2023-04-28 00:00" --until "2023-04-28 08:45" > /tmp/Syslog.log

You may have to change the date/time in the above command.

Can you also post the output of the bellow commands:

ls /etc/pve
ls /etc/pve/nodes
# if the `ls` result that there are files in the above two commands do continue with:
cat /etc/pve/corosync.conf
cat /etc/pve/.members
pveversion -v

qm start 100

 

[thecharliewahl]

Hi,

How did the server disappear? Did you remove it from the cluster?

Can you post the syslog since the issue happened? You can sort the syslog with a specific time/date using journalctl e.g.:

journalctl --since "2023-04-28 00:00" --until "2023-04-28 08:45" > /tmp/Syslog.log

You may have to change the date/time in the above command.

Can you also post the output of the bellow commands:

ls /etc/pve
ls /etc/pve/nodes
# if the `ls` result that there are files in the above two commands do continue with:
cat /etc/pve/corosync.conf
cat /etc/pve/.members
pveversion -v

qm start 100

Hello,

I followed some instructions online on how to remove one from a cluster. ( I have two.)

As for the journalctl it outputs, "/tmp/syslog.log: No such file or directory. When I just run it without date or time it says this, "File is neither a device node, nor regular file, nor executable: /tmp/syslog.log"

 

[Moayad]

Hello,

Thank you for the outputs!

Still the syslog is important to narrow down the issue. However, in the cat /etc/pve/.members shows that the only `saturn` node is in a cluster, and the qm start 100 says that there are no quorm in your cluster. Have you tried to restart the corosync and pve-cluster services?

systemctl restart corosync.service
systemctl restart pve-cluster.service

You can use `pvecm expected 1` [0] command if you want to start the VM.

[0] https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_quorum

 

[thecharliewahl]

I have tried to restart the services. I have now run into a another problem. When I try to log into the shell it says, "TASK ERROR: command '/usr/bin/termproxy 5901 --path /nodes/saturn --perm Sys.Console -- /bin/login -f root' failed: exit code 1"

 

[thecharliewahl]

Is there a way to export the virtual disk so I can keep my data and start a new install?