[SOLVED] Seperate VLAN for management, oddities

[FuriousRage]

Hi.

I have setup vlan 90 on my unify for management.

I setup a vmbr and a vlan on pve network, with vlan 90.

in both servers console i can ping all vlan90 ip's i currently have.

---

However, since i didnt do vlan before i clustered them i did the following on each node:

systemctl stop pve-cluster
systemctl stop corosync

Set the pve and pve2 name to 192.168.90.2 or 90.3
nano /etc/hosts

systemctl start corosync <- starts
systemctl start pve-cluster <- fails

same on both nodes.

journalctl -xeu pve-cluster.service
Jun 05 19:54:11 pve pmxcfs[1128988]: [main] notice: resolved node name 'pve' to '192.168.90.2' for default node IP address
Jun 05 19:54:11 pve pmxcfs[1128988]: [main] notice: resolved node name 'pve' to '192.168.90.2' for default node IP address
Jun 05 19:54:11 pve pmxcfs[1128988]: [main] notice: unable to acquire pmxcfs lock - trying again
Jun 05 19:54:11 pve pmxcfs[1128988]: [main] notice: unable to acquire pmxcfs lock - trying again

and on the webgui, i get an error "no route to host" if i try go to the "other" node from the "first" node.

the interfaces for the vlan looks like
iface vmbr1 inet manual
bridge-ports nic2
bridge-stp off
bridge-fd 0
bridge-vlan-aware yes
bridge-vids 2-4094

auto vmbr1.90
iface vmbr1.90 inet static
address 192.168.90.2/24
#Management VLAN


What detail am i missing here? (old pve only ip was 192.168.88.0/24. want to move the management into a specifc vlan instead of the shared vlan vm/lxc's will have)

 

[FuriousRage]

Ah well, decided to try remove the cluster and have them rejoin.

Backed up the 3 lxc's and deleted the guests (kept backups)

did the whole

systemctl stop pve-clustersystemctl stop pve-cluster
systemctl stop corosync

pmxcfs -l
rm /etc/pve/corosync.conf
rm -r /etc/corosync/*

killall pmxcfs
systemctl start pve-cluster

pvecm delnode pve
pvecm delnode pve2

pvecm expected 1
rm /var/lib/corosync/*

But when i try to join the empty node i get:

detected the following error(s):
* this host already contains virtual guests
TASK ERROR: Check if node may join a cluster failed!

But there is no vm's or lxc's presented in the webgui, and there is no config files in the pve/lxc folder (its completely empty)

 

[FuriousRage]

I found out the reason.


I had to do mkdir /root/nodes && mv /etc/pve/nodes/pve /root/nodes then i could rejoin the cluster on the new ip