Need deamon restart to access the WebGUI

[Sessa45]

I found another strange thing.

lsblk shows:

sda 8:0 1 7,3T 0 disk
├─sda1 8:1 1 1M 0 part
├─sda2 8:2 1 977M 0 part /boot
└─sda3 8:3 1 7,3T 0 part
├─pve-root 253:0 0 190,8G 0 lvm /
├─pve-swap 253:1 0 8192M 0 lvm [SWAP]
└─pve-data 253:3 0 7,1T 0 lvm /var/lib/vz

df -lh shwos:

udev 63G 0 63G 0% /dev
tmpfs 13G 1,3G 12G 11% /run
/dev/mapper/pve-root 187G 14G 164G 8% /
tmpfs 63G 63M 63G 1% /dev/shm
tmpfs 5,0M 4,0K 5,0M 1% /run/lock
tmpfs 63G 0 63G 0% /sys/fs/cgroup
/dev/sda2 946M 351M 531M 40% /boot
/dev/mapper/pve-data 7,1T 962G 5,8T 15% /var/lib/vz

I have checked the HDDs, there are fine.
I seems like i have only 1 TB HDD in use, but pvs shows:

PV VG Fmt Attr PSize PFree
/dev/sda3 pve lvm2 a-- 7,27t 0

And when i list the physical volume it shows with pvdisplay:

--- Physical volume ---
PV Name /dev/sda3
VG Name pve
PV Size 7,27 TiB / not usable 4,00 MiB
Allocatable yes (but full)

df -lh shows enough free space, but pvdisplay does not.
What's this? Could it be that the error is caused by the supposedly non-existent memory? But the monitoring also found no problem. Why does pvdisplay show that the memory is empty?

 

[sb-jw]

You have to understand LVM: https://debian-administration.org/article/410/A_simple_introduction_to_working_with_LVM

In short, it seems okay for me. There is no problem with your Disks or free / used space.

 

[Sessa45]

All right, I have to take a closer look.
At the moment I understood it the same way, that everything is OK. Nevertheless, I did not want to omit any "find" here

The problem is still there. What I find curious is that I can now access the VMs via the GUI and via the console, but the CTs and the storages are still provided with the question mark.

 

[sb-jw]

Could you post your storage Configs here? Maybe there is a typo or other problem visible.

Did you tried it already in another browser?

 

[Sessa45]

Could you post your storage Configs here? Maybe there is a typo or other problem visible.

What exactly do you need?

Did you tried it already in another browser?

Yes, i have already tried this.


It seems that this problem is related to a backup job. The problem occurred at 3 o'clock at night. Exactly at the time when a backup job of a container should be running / running.
I have 2 HDDs running with an ext4 file system in RAID1. Have already performed an unmount, unfortunately, the problem is still present.

 

[sb-jw]

What exactly do you need?

/etc/pve/storage.cfg

Did you already reboot the server?
So you mean at the backup the server goes to an unresponsive state regarding the storage?

 

[Sessa45]

Did you already reboot the server?

No. It is productive.
But the server itself is running fine.

/etc/pve/storage.cfg

dir: local
path /var/lib/vz
content iso,vztmpl,rootdir,images
maxfiles 5
shared 0

dir: ssd
path /ssd_lvm
content iso,vztmpl,rootdir,images
shared 0

dir: backup_storage
path /backup_storage
content backup
maxfiles 100
shared 0

Every VM shows this graph (related to the timestamp)
So from this time the connection get lost i think.

 

[sb-jw]

Try to restart the service or check the status before restart: https://pve.proxmox.com/pve-docs/pvestatd.8.html

Then check if the stats are available again and if they failed again when the backup are running.

 

[Sessa45]

I already restarted the services as i mentioned.

The problem is still there.

service pvedaemon restart
service pveproxy restart
service pvestatd restart

Then i can browse a while throug the interface. KVM is visible but lxc has a question mark on the icon in the server list.

 

[Sessa45]

It seems like the node is running as cluster configuration.
How can i set up / change to single node?

 

[Stoiko Ivanov]

Are there other nodes in the cluster? (`pvecm status` should provide that insight)

The cleanest way to create a single node would be:
* backup all VMs
* install PVE fresh
* restore from backup

Apart from that you should be able to change it to a single node by following our reference documentation:
https://pve.proxmox.com/pve-docs/pve-admin-guide.html#chapter_pvecm
(more specifically https://pve.proxmox.com/pve-docs/pve-admin-guide.html#_remove_a_cluster_node)