Massive headache after moving mini pc to another room

[50p]

Hi,
I'm very new to proxmox. I've played around with it on a VM before I actually went and installed it on a machine.
I installed PVE on a mini PC and configured it the way I liked. I installed Open Media Vault to be able to share files to all my devices using SMB and mounted external 4TB HDD (most of machines are Windows but can access files on Android phone with Cx File Explorer). I've tried nextcloud but the upload/download speeds weren't satisfying (upload speed ~500KB/s) so I went for OMV. I still run nextcloud, but don't use it. I was happy with the OMV setup but again, wasn't happy with the speed (upload speed ~20MB/s, on 1Gb/s lan). I know the external HDD is capable of at least 120MB/s because I tested it on my main desktop, I would be happy with 50MB/s. The mini PC was in a room where ethernet was delivered via powerline adapter (obviously slower than direct connection to router). So, I decided to move the min PC and connect it to the router directly. So, I've shutdown OMV VM and then shutdown PVE node. It's a single-node setup.
When I moved it to the other room, it worked fine but the speed dropped... upload speed 10-13MB/s. The next day I started to get issues logging in to web interface. Eventually when logged in, I couldn't see any nodes/CTs/VMs. Open Media Vault worked and I could log in to its web interface but I couldn't see anything on proxmox web UI.
I started trying to figure out why it's problematic and asked ChatGPT. It's taken me 2 days now trying to figure out why I have the problem but ChatGPT just isn't helping. I'm spinning in circle.
Currently, I can see the nodes on proxmox web interface but I can't change any settings and I can see message "Error Connection error - server offline?" in Notes box of nextcloud container and OVM VM. The Status box shows labels for CPU and Memory usage but it's grayed out and I can see a spinning wheel with "Connection error" on top of that.
I can still access the files/folder on SMB and I can log in to OVM web interface so the VM runs but I can't change anything or create new containers/VMs on proxmox.
I have to ssh to proxmox to be able to do anything. I can't access Shell or Console via proxmox web interface.

pvecm status

Cluster information
-------------------
Name:             UNKOWN?
Config Version:   -1
Transport:        knet
Secure auth:      off

Quorum information
------------------
Date:             Fri Jan 17 20:34:20 2025
Quorum provider:  corosync_votequorum
Nodes:            1
Node ID:          0x00000001
Ring ID:          1.3c
Quorate:          Yes

Votequorum information
----------------------
Expected votes:   1
Highest expected: 1
Total votes:      1
Quorum:           1
Flags:            Quorate

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 192.168.0.111 (local)

With the help of ChatGPT I needed to configure corosync.conf.

totem {
    version: 2
    secauth: off
    interface {
        ringnumber: 0
        bindnetaddr: 192.168.0.111
    }
}

nodelist {
    node {
        name: pve
        nodeid: 1
        quorum_votes: 1
        ring0_addr: 192.168.0.111
    }
}

quorum {
    provider: corosync_votequorum
    expected_votes: 1
}

logging {
    fileline: off
    to_syslog: yes
}

● pmxcfs.service - Proxmox Cluster File System (pmxcfs)
     Loaded: loaded (/etc/systemd/system/pmxcfs.service; enabled; preset: enabled)
     Active: active (running) since Fri 2025-01-17 19:50:45 GMT; 51min ago
    Process: 35682 ExecStart=/usr/bin/pmxcfs (code=exited, status=0/SUCCESS)
   Main PID: 35694 (pmxcfs)
      Tasks: 5 (limit: 18930)
     Memory: 16.0M
        CPU: 1.278s
     CGroup: /system.slice/pmxcfs.service
             └─35694 /usr/bin/pmxcfs

Jan 17 19:51:02 pve pmxcfs[35694]: [confdb] crit: cmap_initialize failed: 2
Jan 17 19:51:02 pve pmxcfs[35694]: [dcdb] crit: cpg_initialize failed: 2
Jan 17 19:51:02 pve pmxcfs[35694]: [status] crit: cpg_initialize failed: 2
Jan 17 19:51:08 pve pmxcfs[35694]: [confdb] crit: cmap_get totem.config_version failed 12
Jan 17 19:51:08 pve pmxcfs[35694]: [confdb] crit: cmap_get totem.cluster_name failed 12
Jan 17 19:51:08 pve pmxcfs[35694]: [status] notice: node has quorum
Jan 17 19:51:08 pve pmxcfs[35694]: [dcdb] notice: members: 1/35694
Jan 17 19:51:08 pve pmxcfs[35694]: [dcdb] notice: all data is up to date
Jan 17 19:51:08 pve pmxcfs[35694]: [status] notice: members: 1/35694
Jan 17 19:51:08 pve pmxcfs[35694]: [status] notice: all data is up to date

If there is anything else you may need to help me out.
Like I said I'm very new to this and I'm not sure what troubleshooting steps I should take. I kept asking ChatGPT and giving it results but I feel like I'm not getting anywhere.
Did I corrupt some files when I moved the server to another room?

Thank you for any help. I don't normally ask for help because I like to solve problems myself but I am frustrated and I know proxmox community is very helpful and hopefully you can help.
I would like to avoid reinstalling proxmox as well because I installed NordVPN on OVM. There is a limit of 10 devices for NordVPN account and I'm not sure if I reinstall proxmox, that 1 device will be used but inaccessible. Also, I've already transferred 84GB of data to the external HDD, and I'm not sure if they will be accessible once I reinstall proxmox and a new VM for OMV

 

[leesteken]

When I moved it to the other room, it worked fine but the speed dropped... upload speed 10-13MB/s.

Sounds like a 100Mbps cable (=12.5MB/s) or an unreliable cable or port where 1Gbps speeds could not be negotiated. Did you try another cable and/or check on the router side what the connection type is like? Did you try moving the system back to see if the problem goes away?

The next day I started to get issues logging in to web interface. Eventually when logged in, I couldn't see any nodes/CTs/VMs. Open Media Vault worked and I could log in to its web interface but I couldn't see anything on proxmox web UI.

That might be unrelated and/or a different problem all together. Did you try moving the system back to see if this problem stays while the network connection gets better?

The whole ChatGPT and corosync editing seems like a bad idea (as you are not running a cluster) and might have made things worse. Maybe reinstall Proxmox and restore your VM/CTs from backups?

 

[50p]

Thank you for quick reply.

I haven't tried moving it back to the other room but it may be risky now because the VM and CT don't always run after reboot.

I haven't backed up VM nor CT. What commands do I use to properly back them up and restore after reinstallatio? Is it a longer process that I should follow a YouTube tutorial or ChatGPT? I will probably do it tomorrow, I have had enough of this for today.

Thanks again.

Edit:
The cable I use is 5e. I will try maybe different port on route, but I don't think there would be 100Mbps port if other 3 ports are used and I don't notice slow speeds on them (TV, laptops, etc.)

 

[BobhWasatch]

One broken wire in the cable will force the link to 100 Mbit/s.

 

[50p]

One broken wire in the cable will force the link to 100 Mbit/s.

Thanks, I was thinking of testing a different cable. Even though it's 5e, it feels thin.

 

[Kingneutron]

A) You should be asking human experts before chatgpt, it's known to give shite advice and make things up out of thin air.

B) YOU NEED BACKUPS. Plan to setup Proxmox Backup Server somewhere so you get dedup + other features.

This also kind of sounds like you might be using crappy consumer-level drives, what are the make/model/size of drives that you have Proxmox installed to / use for data? SMR might be killing your performance, and Heaven help you if you're using something like a BX500 or QLC SSD. Both are absolute dumpster-fire level crappy tech

How you have things configured on the virtual side also matters, you should be using virtio networking with multiqueue and not over-allocating CPU and RAM to VMs. Virtual disk config also matters, if you're using virtual IDE then you're gonna have a bad time.

 

[50p]

I use WD 256GB SSD NVMe for the system and Ironwolf NAS 4TB HDD for storage. I might attach an external Kingston A400 480GB SSD for backups which I've used as extra storage a few years ago.
I have an HP 400 G4 mini PC (i5 8500T, 16GB RAM) and only run a VM for Open Media Vault and a container for nextcloud (I will probably not bother with it), some test node (e.g. Wireguard) that I can ignore for now.
So, I have planty of power and storage to run a couple of containers and a VM.

I might have a look at backing up the VM and a container tonight. I will need to make sure I will do it properly so that once I restore the VM/CT, I want them to use the same HDD for storage and OMV to detect the data I've already uploaded to it. Is it just a matter of mounting the HDD to the same folder as it is now (e.g. /mnt/usb-storage)?

Thanks.