PVE Recovery

HamJam05

New Member
Mar 2, 2025
10
0
1
I moved recently and now have a new ISP. I am unable to connect to my pve and need help getting back to the environment. I have tried messing around with the CLI and switching IPs and whatnot. Ultimately I need the help of someone smarter than me :)

I've changed my gateway IP and host IP. I used these commands to do so, 'nano /etc/network/interfaces' & 'nano /etc/hosts'

Please help, I'm stressing over here.
 
There are a few files which may be affected too, see "grep -r <my-IP> /etc" which you all have to consider and update.
 
No, look for your old ip to find out where it is defined to know what to change to new one.
 
No, look for your old ip to find out where it is defined to know what to change to new one.
just tried that as well. still says no such file/ directory exists.

this is why I need the help, I don't mean to be confusing or pestering.

are you on discord? if so, when convenient for you, we could use that platform to speed this process
 
These files are the results of grep:
/etc/resolv.conf:nameserver 192.168.50.253
/etc/hosts
/etc/issue
/etc/pve/.members
/etc/pve/priv/known_hosts
/etc/ssh/ssh_known_hosts
/etc/pve/corosync.conf
/etc/corosync/corosync.conf # This is auto-updated of changes are made in /etc/pve/corosync.conf
/etc/network/interfaces
/etc/pve/nodes/*/lxc/*.conf # Maybe if have static ip in lxc ...
/etc/pve/nodes/*/qemu-server/*.conf # Maybe if have static ip in vm ...
 
i think i did it correctly this time. i had to run 'nano /etc/network/interfaces' & 'nano /etc/hosts' to change my new IP back to the old one before I moved. Once I did that, I restarted the server, then ran the grep. I included the results in a picture below. I'm not worried about showing IP since its the old one and I need to update it so it can work again
 

Attachments

  • IMG_0067.jpg
    IMG_0067.jpg
    498.3 KB · Views: 8
What is the full output of?:
  • ip a
  • cat /etc/network/interfaces
  • cat /etc/hosts
And what is your actual subnet? (Check e.g. your router and/or working client.)
 
What is the full output of?:
  • ip a
  • cat /etc/network/interfaces
  • cat /etc/hosts
And what is your actual subnet? (Check e.g. your router and/or working client.)
here are the results of the commands you listed above. also, this was done with my old IP details plugged in. the subnet mask is /24 (255.255.255.0)
 

Attachments

  • IMG_0071.jpg
    IMG_0071.jpg
    990.4 KB · Views: 9
  1. ip a output is missing
  2. /etc/hosts must not contain the/a port (:8006)
  3. What is your actual IP address range you are using, is what I meant. E.g. what is the IP each of your router/gateway and a working client?

FYI:
I'm not worried about showing IP since its the old one [...]

Read the following for why you do not need to worry at all, if you are using private IP addresses / address ranges...:
https://en.wikipedia.org/wiki/Private_network
 
1. sorry, i included the ip a output
2. I ran nano /etc/hosts to remove the port, :8006
3. The current IPs:
  • Router/ Gateway- 192.168.4.1
  • Client- 192.168.4.23 (the computer I'm on now)
3.1 The old IPs used before we moved:
  • Router/ Gateway- 10.0.0.1
  • Client- 10.0.0.182/24
 
sorry, the photo was too large in the last message
  1. ip a output is missing
  2. /etc/hosts must not contain the/a port (:8006)
  3. What is your actual IP address range you are using, is what I meant. E.g. what is the IP each of your router/gateway and a working client?

FYI:


Read the following for why you do not need to worry at all, if you are using private IP addresses / address ranges...:
https://en.wikipedia.org/wiki/Private_network
 

Attachments

  • IMG_0085.JPG
    IMG_0085.JPG
    228.2 KB · Views: 5
Why do you omit the full output by taking vmbr0 down before?

However:
  1. Replace the: 10.0.0.182 in: /etc/network/interfaces and: /etc/hosts with a free IP address of your liking (192.168.4.XY) from within the: 192.168.4.0/24 subnet, but outside the DHCP-range of your router.
  2. Replace the: 10.0.0.1 in: /etc/network/interfaces with: 192.168.4.1.
  3. Reboot the PVE-host.

Test/Troubleshooting:
  • Try to open the webUI (https://192.168.4.XY:8006) on a working client.
  • Try to ping the PVE-host (192.168.4.XY) from a working client.
  • Try to ping the router (192.168.4.1) from the PVE-host.

If it is not working, report the results from the above tests here and provide the full output of: ip a again.

PS.: Admittedly, I am unsure about that FQDN in your: /etc/hosts...
 
Real quick I just want to say thank you guys for the volunteered help. Im not as well versed in this as others so what may seem or be simple to you is crucial learning for me. That being said I'm providing as much information as I possibly can to help you guys help me, so thank you guys. :)


  1. Inserted the IP 192.168.4.240into: /etc/network/interfaces and: /etc/hosts.
  2. Inserted the new gateway IP; 192.168.4.1 into: /etc/network/interfaces.
  3. Rebooted.

Results of said actions.

Before we get into the pinging, I noticed on my router software/ app, that the proxmox server and the containers I have in there are showing "online."
In the app, my 'adguard' container is assigned with IP 192.168.4.28
'casaos' container is assigned IP 192.168.4.27
Here is what is confusing me 'proxmox' is being assigned 192.168.4.31however, I used the CLI to assign it 192.168.4.240


1741155418651.png1741155429442.png

Regardless, I cannot remote into any of these IP addresses. I have tried multiple times and it makes me want to pull my hair out!



Test/Troubleshooting:

  • Try to ping the PVE-host (192.168.4.XY) from a working client. I can ping both IPs 192.168.4.31 & 192.168.4.240, which are assigned to the PVE host. I used my main computer to ping them. For both IPs, 4 packets sent and received with 0 delays and loss percentage.

  • Try to ping the router (192.168.4.1) from the PVE-host. I can successfully ping the router/ gateway from the PVE host. Sent 9 packets and received 9 before ending the ping test. No discrepancies.

Here are the results of the IP a
1741156984686.png

Im so lost:(
 
I used /etc/network/interfacesto switch the subnet mask from /32 to /24. That did not work unfortunately. I'm thinking of backing up the containers and reinstalling. I just really wanted to avoid that at the risk of losing data.