What exactly does the migrator think it is doing?

[AngryAdm]

It takes half a day to copy Disk0.
Then it copies the EFI disk which is fine.
Then it realizes it has already create Disk0 and decides to create disk2 and format it again.... Why?
There is no disk2on the host... There was Disk0 1TB and Disk1 an EFI disk...


Task viewer: VM 110 - Migrate

OutputStatus

Stop
2020-10-02 12:02:47 starting migration of VM 110 to node 'pve03' (10.0.0.7)
2020-10-02 12:02:47 found local disk 'pve01-vmstore:110/vm-110-disk-0.qcow2' (in current VM config)
2020-10-02 12:02:47 found local disk 'pve01-vmstore:110/vm-110-disk-1.qcow2' (in current VM config)
2020-10-02 12:02:47 found local disk 'pve02-vmstore:110/vm-110-disk-0.qcow2' (via storage)
2020-10-02 12:02:47 found local disk 'pve02-vmstore:110/vm-110-disk-1.qcow2' (via storage)
2020-10-02 12:02:47 copying local disk images
2020-10-02 12:02:49 Formatting '/storage/vm/images/110/vm-110-disk-0.qcow2', fmt=qcow2 cluster_size=65536 preallocation=metadata compression_type=zlib size=1073741824000 lazy_refcounts=off refcount_bits=16
2020-10-02 17:39:25 262184064+0 records in
2020-10-02 17:39:25 262184064+0 records out
2020-10-02 17:39:25 1073905926144 bytes (1.1 TB, 1000 GiB) copied, 20195.5 s, 53.2 MB/s
2020-10-02 17:39:44 3455+104699206 records in
2020-10-02 17:39:44 3455+104699206 records out
2020-10-02 17:39:44 1073905926144 bytes (1.1 TB, 1000 GiB) copied, 20036 s, 53.6 MB/s
2020-10-02 17:39:44 successfully imported 'pve02-vmstore:110/vm-110-disk-0.qcow2'
2020-10-02 17:39:45 volume 'pve02-vmstore:110/vm-110-disk-0.qcow2' is 'pve02-vmstore:110/vm-110-disk-0.qcow2' on the target
2020-10-02 17:39:49 112+0 records in
2020-10-02 17:39:49 112+0 records out
2020-10-02 17:39:49 458752 bytes (459 kB, 448 KiB) copied, 0.0228926 s, 20.0 MB/s
2020-10-02 17:39:49 Formatting '/storage/vm/images/110/vm-110-disk-1.qcow2', fmt=qcow2 cluster_size=65536 preallocation=metadata compression_type=zlib size=131072 lazy_refcounts=off refcount_bits=16
2020-10-02 17:39:49 7+0 records in
2020-10-02 17:39:49 7+0 records out
2020-10-02 17:39:49 458752 bytes (459 kB, 448 KiB) copied, 0.000649331 s, 706 MB/s
2020-10-02 17:39:49 successfully imported 'pve01-vmstore:110/vm-110-disk-1.qcow2'
2020-10-02 17:39:49 volume 'pve01-vmstore:110/vm-110-disk-1.qcow2' is 'pve01-vmstore:110/vm-110-disk-1.qcow2' on the target
2020-10-02 17:39:50 file '/storage/vm/images/110/vm-110-disk-0.qcow2' already exists - importing with a different name
2020-10-02 17:39:50 Formatting '/storage/vm/images/110/vm-110-disk-2.qcow2', fmt=qcow2 cluster_size=65536 preallocation=metadata compression_type=zlib size=1073741824000 lazy_refcounts=off refcount_bits=16

 

[AngryAdm]

I stopped the creation of disk "2".
Created a new vm on the target vm and mounted the disks it already had copied. Mission complete.
No reason to wait for another 6 hours while it does illogical stuff..

 

[fabian]

it tells you exactly what it's doing:

2020-10-02 12:02:47 found local disk 'pve01-vmstore:110/vm-110-disk-0.qcow2' (in current VM config)
2020-10-02 12:02:47 found local disk 'pve01-vmstore:110/vm-110-disk-1.qcow2' (in current VM config)
2020-10-02 12:02:47 found local disk 'pve02-vmstore:110/vm-110-disk-0.qcow2' (via storage)
2020-10-02 12:02:47 found local disk 'pve02-vmstore:110/vm-110-disk-1.qcow2' (via storage)
2020-10-02 12:02:47 copying local disk images

it found 4 disks, and copied them.

maybe your storage configuration/visibility is wrong? are pve01 and pve02 actually the same (shared) storage?

 

[AngryAdm]

No, they are two different hosts. There was only 2 disk files in the /vm/images/110 dir. pve02 did not contain the disks physically.
They do not share storage. I suspect I misconfigured a datastore to be avail from both hosts.