Proxmox main node wrong IP in certificates

[harethtvalomer]

Hello everyone, I noticed something when I try to add another node to the cluster,
When I paste the join information it shows the wrong IP But that's not issue for now since I can uncheck assisted join and change the IP

Now when I go to my main node -> certificates, I see the pve-ssl.pem have the wrong IP
I have no Idea how to fix it, I tried generating certificate with --force, It does not work

 

[dcsapak]

can you post your /etc/network/interfaces as well as you /etc/hosts ?

 

[dcsapak]

also /etc/pve/corosync.conf

 

[harethtvalomer]

also /etc/pve/corosync.conf

logging {
  debug: off
  to_syslog: yes
}

nodelist {
  node {
    name: 25253
    nodeid: 1
    quorum_votes: 1
    ring0_addr: #CORRECT_IP
  }
  node {
    name: g1
    nodeid: 2
    quorum_votes: 1
    ring0_addr: #CORRECT_IP
  }
  node {
    name: nlb1
    nodeid: 4
    quorum_votes: 1
    ring0_addr: #CORRECT_IP
  }
  node {
    name: usa2
    nodeid: 3
    quorum_votes: 1
    ring0_addr: #CORRECT_IP
  }
}

quorum {
  provider: corosync_votequorum
}

totem {
  cluster_name: usa-node
  config_version: 20
  interface {
    linknumber: 0
  }
  ip_version: ipv4
  link_mode: passive
  secauth: on
  version: 2
}

 

[harethtvalomer]

can you post your /etc/network/interfaces as well as you /etc/hosts ?

127.0.0.1       localhost
31.207.44.186   srv42278
#SAME_IP_AS_COROSYNC_NODE    25253.example.com       25253

# The following lines are desirable for IPv6 capable hosts
::1     localhost ip6-localhost ip6-loopback
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
#SAME_IP_AS_COROSYNC_NODE    25253.example.com       25253

auto lo
iface lo inet loopback

iface enp2s0f0 inet manual

iface enp2s0f1 inet manual

auto vmbr0
iface vmbr0 inet static
        address #SAME_IP_AS_COROSYNC/25
        gateway #GATEWAY
        bridge-ports enp2s0f1
        bridge-stp off
        bridge-fd 0
        bridge_hw 0c:c4:7a:18:5e:b7

 

[dcsapak]

ok so your hostname is numbers only, so it gets interpreted by glibc directly as an ipv4 address (a number is also a valid ip address)
i'm afraid there is no easy fix for that right now besides using at least one non-digit character in the nodename

 

[harethtvalomer]

ok so your hostname is numbers only, so it gets interpreted by glibc directly as an ipv4 address (a number is also a valid ip address)
i'm afraid there is no easy fix for that right now besides using at least one non-digit character in the nodename

Okay, Can you link me a doc where I can rename the hostname without screwing the node? Since I have customer's data on it.

 

[dcsapak]

well editing a nodename on a node with existing vms and data is not really recommended or tested procedure.
it would be better to migrate the vms off the node, remove it from the cluster and begin new with that node
(the reason being that the guest configs and some other stuff depends on the nodename (=hostname) in the cluster filesystem/config)

if you know what you are doing, it probably is possible (renaming the node in /etc/hosts, /etc/hostname, adapting the cluster config, moving the node directory in /etc/pve, etc...)
but still not recommended