HOW to assign NAS located vms from one node (that crashed) to another node?

[philjans]

Hi,
I have a cluster of 5 nodes. The 5th one is not working (because it was upgraded and now don't play nice with the rest).
Untill I fix the 5th node: I want to access my vm which are located on a FreeNAS using any of the other 4 nodes in the cluster?
That should be simple, right?



Thanks for the help

p.s. details about the crash node 15 and trying to downgrade it are located here: https://forum.proxmox.com/threads/p...o-not-work-in-the-cluster-anyone-knows.78219/

 

[AKA]

There is so much information missing here.
Which storage are you using? Local ZFS?
Or is the ZFS only used for boot and you're using CEPH or gluster or another Shared Storage?

Do you have HA configured?

 

[philjans]

I understand. My proxmox admin left and I'm a vmware admin.
I don't know where the actual promox OS is located, probably local ssd on the blade itself but all vm are located on a FreeNas.
CEPH is not installed.

Here's what I see for the storage:
HA:


How can I find the other piece of information you need?
tx

 

[AKA]

Your fix should be simple then.

You can have access to the VMs moving the files in /etc/pve/nodes/DEAD NODE/qemu-server/*.conf to another working node in the same directory.
Then you can start those VMs on another node. HA would've done that for you if it was configured (Which i don't see any reason why it shouldn't.

BUT!
I'm still confused. In your first screenshot pvebld15 was down.
In the screenshot you posted just not pvebld05 seems to be. I don't quite understand that.

 

[philjans]

Yes the initial issue was with pvebld15 that was down.
But I was able to have it worked.
Today it was pvebld05 what was down. A simple reboot and it worked.
Tomorrow I will have all 5 of them upgraded to the same version which could raise stability.

Yes I don't know what is wrong with HA... missing configs probably.

I'll look into your "/etc/pve/nodes/DEAD NODE/qemu-server/*.conf". thanks for this information

 

[AKA]

Yes, you need confiugration.

Under HA => Resources => Add

You can add VMs there so that they get automatically moved when a hardware node crashes.

You can find something here: https://pve.proxmox.com/wiki/High_Availability
It describes mostly command line functions, but alos a general understanding.
You should also look into the values: max relocate and max restart before adding VM's, since you have 5 nodes.