[SOLVED] QM Migrate Confusion

[proxdrop]

qm migrate is a simple enough command - I just need to specify the VMID and node, and it should just work. However, I can only "push" VMs to remote nodes - I can't "pull" vms to my current node. Why is this?

Syntax/Situation:
3 node cluster, ceph enabled and accessible on all. vm1234 resides on ceph
node1
node2 - working directory
node3 - contains vmid 1234

qm migrate 1234 node2 --online fails: 400 parameter verification failed. target: node is local node.

What gives? Moving via web gui and moving from node2 -> node3 all work just fine. Is the only "alternative" to ssh into node3 to "push" VMID 1234 back to node2? Is there an alternate pvesh command with similar functionality?

EDIT: Is this a consequence of the vm config file "living" on the remote server, not the local?

 

[bbgeek17]

Hi @proxdrop ,

"qm" command interface can only manipulate the VMs local to where "qm" is running. Lets just say for historical reasons.

If you want to "pull" migrate (or do anything else), you need to use pvesh interface. There you can and must specify the Node you are addressing.
You can also use API directly, of course.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[proxdrop]

Hi @proxdrop ,

"qm" command interface can only manipulate the VMs local to where "qm" is running. Lets just say for historical reasons.

If you want to "pull" migrate (or do anything else), you need to use pvesh interface. There you can and must specify the Node you are addressing.
You can also use API directly, of course.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

Thanks. Hilariously, in my research I found your old post on the matter and the syntax worked, so thanks.

pvesh create /nodes/"$start_host"/qemu/"$vmid"/migrate -target "$end_host" -online 1

https://forum.proxmox.com/threads/m...nother-within-same-cluster.129195/post-565669