The PVE can't boot up and stuck on "Loading Initial ramdisk …"

[shore]

Hi!
I am new to PVE, and I am facing a problem that I don't know how to solve, I hope to get some help, thx!
My computer's PVE can't boot up after moving to another location (it was working before), every time I try to boot up PVE, it gets stuck with this:

Loading Linux 6.5.13-1-pve …
Loading Initial ramdisk …

I try to selecting Proxmox VE GNU/Linux, with Linux 6.5.13-1-pve (recovery mode) from the boot menu, and it stucked on this screen:


I think the problem might be with the hard disk, but I'm not sure. Can anyone please help?

 

[gfngfn256]

My computer's PVE can't boot up after moving to another location (it was working before)

Well, if really nothing else changed, except the server was physically moved, then check all cables are correctly connected/inserted, & check all internal disks are correctly seated/connected in their slots/connectors. Also make sure all RAM is perfectly seated. Something probably came loose while transporting.

 

[milew]

It doesn't necessarily have to be a local device.
Start the host in single mode
You can look at the entries in /etc/fstab
and the logs

journalctl -u sys-fs-fuse-connection.mount

 

[shore]

Well, if really nothing else changed, except the server was physically moved, then check all cables are correctly connected/inserted, & check all internal disks are correctly seated/connected in their slots/connectors. Also make sure all RAM is perfectly seated. Something probably came loose while transporting.

Thank you. I will check it

 

[shore]

This host is part of the cluster. The other host is still working, but because this host is offline, the shell doesn't work. Is the problem related to the cluster?
And I start with the "nomodeset" parameters, and it stuck on this right now:

 

[gfngfn256]

So the "location" (physical) of the server changed. Now we learn it is part of a cluster. Has the NW changed? Do you have routers/switches/cables exactly as they were initially?

 

[shore]

So the "location" (physical) of the server changed. Now we learn it is part of a cluster. Has the NW changed? Do you have routers/switches/cables exactly as they were initially?

The network has changed.
And there are two hosts in this cluster. The first node is still working, but the second node, which is the host we are talking about, can't boot up anymore. I moved the entire cluster (both hosts) to my friend's house. Now, the IP addresses of the fisrt node (still work) is assigned as private IP (like 192.168.1.xxx) by DHCP, with its gateway and DNS server addresses set to the IP address of the router of the house. I can't know the IP of the second node because it can't boot up. However, I login the WebUI of the first node and the cluster menu display the old IP (before move) and I can't change anything because the second node is offline.
The second node not being able to connect was the initial problem, and I thought that after solving the problem of the second node not being able to boot up (which is what I'm asking now), the initial problem would be solved.
Could the network problem be causing the boot-up error? Did the network change lead to the boot-up problem, or did the boot-up problem lead to the disconnection?

 

[sw-omit]

If it doesn't boot (fully), then it of course doesn't have network nor access to the other server, so lets focus on that first and "ignore" the second server for now.
After we fix the boot-up issue though, we'll also will probably have to fix the cluster, as because of the network-space change, the servers will not be able to find eachother most likely (one of the reasons I always work with static IPs), as well as that because you're in an even-numbered cluster without Q-Device, things are possibly/probably out-of-sync.

Before you moved the server, when was the last time it was completely restarted, for either updates or other maintenance?
And how critical are the VM's / Containers that are on this server + do you have (recent/known-working) backups of them?

 

[gfngfn256]

Now, the IP addresses of the fisrt node (still work) is assigned as private IP (like 192.168.1.xxx) by DHCP

For Proxmox in general & definitely in a cluster you must use a static IP.

 

[shore]

If it doesn't boot (fully), then it of course doesn't have network nor access to the other server, so lets focus on that first and "ignore" the second server for now.
After we fix the boot-up issue though, we'll also will probably have to fix the cluster, as because of the network-space change, the servers will not be able to find eachother most likely (one of the reasons I always work with static IPs), as well as that because you're in an even-numbered cluster without Q-Device, things are possibly/probably out-of-sync.

Before you moved the server, when was the last time it was completely restarted, for either updates or other maintenance?
And how critical are the VM's / Containers that are on this server + do you have (recent/known-working) backups of them?

I'm not sure, the last reboot was about two month ago. There are some data in that error host, that's why I've faced this problem for a month. However, I am think of re-install the whole system now.