Re: [PVE-User] Moving virtual machine from CE 6.X to latest enterprise edition, advice requested

[Peppo Brambilla <peppo.brambilla@unibe.ch>]

Hi

You have to configured the NFS share to allow "disk image" and 
"container" as content.
On PVE 7.4 you have to configure the 'Content' setting when editing the 
storage. If you don't set this up correctly, then also the cli won't do 
the trick.

Cheers -- Peppo

On 04.08.23 11:06, Joseph John wrote:
> Hi ,
> Thanks for the advice
> I tried first with a CIFS share , checked by accessing it from other
> machines . created CIFS share (Left hand side of menu, Data Center,
> Storage) ,
> Now when I go to the VM , backup option , and click backup I can only see
> the local storage, I am not able to see the CIFS share which I created
>
> Similarly I did with NFS server and shared directory , checked it by
> accessing it from other hosts. Created NFS share (Left hand side of menu,
> Data Center, Storage).
> This also when I go the the VM, backup option and click back I can only see
> the local storage , not the NFS or the CIFS storage which we created
> earlier . The prox mox version is 6.3-3.
>
> But when I give a df-h , I get the results
>
> oot@server-1:~# df -h
> Filesystem                               Size  Used Avail Use% Mounted on
> udev                                     378G     0  378G   0% /dev
> tmpfs                                     76G   13M   76G   1% /run
> /dev/mapper/pve-root                      29G  6.2G   22G  23% /
> tmpfs                                    378G   69M  378G   1% /dev/shm
> tmpfs                                    5.0M     0  5.0M   0% /run/lock
> tmpfs                                    378G     0  378G   0%
> /sys/fs/cgroup
> /dev/sdb2                                511M  312K  511M   1% /boot/efi
> /dev/fuse                                 30M   72K   30M   1% /etc/pve
>
> *//10.115.129.160/proxmox <http://10.115.129.160/proxmox>
> 1.0T  359G  666G  36%
> /mnt/pve/backup10.115.129.169:/home/itsupport/nfsshare 1017G   14G  955G
> 2% /mnt/pve/nfsbackup*
> tmpfs                                     76G     0   76G   0% /run/user/0
>
>
> So from the df -h , we can see the CIFS and NFS, but through the web
> interfaces, backup option do not show them in the drop down menu.
> I am thinking in this situation, to make use command line and run the
> backup so that I can specify the location (cifs or nfs share) to which I
> can  copy the disk image
> I will be posting another thread on requesting on how to use command line
> to  take backup of VM images
>
>
>
> Thanks
> Joseph John
>
>
>
> On Thu, Aug 3, 2023 at 7:14 PM Peppo Brambilla <peppo.brambilla@unibe.ch>
> wrote:
>
>> Hi
>>
>> I was using an NFS server connected to the old and new cluster. This
>> should be setup as similar as possible on both cluster. Maybe it's
>> sufficient if they are named the same.
>> Then I proceeded as follows:
>>
>> * move VM's disks to the NFS share
>> * shutdown VM on old cluster
>> * copy VM's config files in /etc/pve/qemu-server/  from old to new server
>> * you should now see the VM on the new server as well
>> * start VM on new server
>> * move disks of VM to ceph on new server (you may want to keep the old
>> disk images)
>> * if the VM runs fine, you can remove the config files on the old
>> cluster and remove the disk images there as well)
>>
>> Cheers -- Peppo
>>
>>
>> On 03.08.23 11:19, Joseph John wrote:
>>> Hi
>>> Thanks.
>>>
>>> The backup option for some reason not completing successfully
>>> from the log files
>>> cat  /var/lib/vz/dump/vzdump-qemu-109-2023_08_02-12_34_08.log
>>> 2023-08-02 12:42:10 ERROR: vma_queue_write: write error - Broken pipe
>>> 2023-08-02 12:42:10 INFO: aborting backup job
>>> 2023-08-02 12:42:10 INFO: stopping kvm after backup task
>>> 2023-08-02 12:42:13 ERROR: Backup of VM 109 failed - vma_queue_write:
>> write
>>> error - Broken pipe
>>>
>>> the GUI /Web way of taking backup not working
>>> I have to find another way of taking the image and putting it on the new
>>> setup , the present one storage is based on the ceph storage and the vm
>> are
>>> in the ceph pook
>>> I am thinking a way in which I can try to take to VM image from the ceph
>>> pool  and move it to the other  new pve instance
>>>
>>> Guidance requested  for getting the vm images to new proxmox
>>>
>>>
>>>
>>> Thanks
>>> Joseph John
>>>
>>>
>>>
>>> On Wed, Aug 2, 2023 at 11:37 AM Aaron Lauterer <a.lauterer@proxmox.com>
>>> wrote:
>>>
>>>> That sounds like a reasonable plan.
>>>>
>>>> If you have a network share that you want to utilize for the backups
>> later
>>>> on
>>>> (never store them on the same machine, 3-2-1 backup strategy and such
>> ;) )
>>>> you
>>>> could already configure it on the old Proxmox VE server. That way, you
>>>> don't
>>>> have to manually move the backup files over. Since you are
>> decommissioning
>>>> the
>>>> old server, there is no risk of running into issues of two different
>>>> Proxmox VE
>>>> clusters (or separate single nodes), accessing the exact same storage.
>>>>
>>>> Under normal situations, you don't want to give two Proxmox VE clusters
>>>> access
>>>> to the same storage, as you might run into VMID conflicts.
>>>>
>>>> Cheers,
>>>> Aaron
>>>>
>>>> On 8/2/23 09:31, Joseph John wrote:
>>>>> Dear All,
>>>>> Good morning
>>>>> We are going to have ProxMox 8 enterprise edition setup, installation
>> is
>>>>> going on.
>>>>>
>>>>> Earlier we were using ProxMox 6.X Community Edition, this is a
>> separate
>>>>> unit which we were using . Now once the enterprise edition is up, we
>> plan
>>>>> to move all the VM instance which were on the old 6.X  CE to the
>>>> separate
>>>>> ProxMox 8 enterprise edition
>>>>>
>>>>> I am planning to move the old vm form the CE (6.X) to ProxMox 8
>>>> (Enterprise
>>>>> Edition)  in the following way
>>>>>
>>>>>       - Take  backup of the VM at CE using the backup option
>>>>>       - scp/rsync  the  backup files to the Prox Mox 8 Enterprise
>> version
>>>>>       - Use the restore option from the new server to restore the image
>>>>>
>>>>>
>>>>> Like to get advice, if the above mentioned steps is apt when you are
>>>> going
>>>>> to move from one of the older CE edition [6.X] to the latest enterprise
>>>>> edition
>>>>>
>>>>> Thanks
>>>>> Joseph John
>>>>> _______________________________________________
>>>>> pve-user mailing list
>>>>> pve-user@lists.proxmox.com
>>>>> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user
>>>>>
>>>>>
>>> _______________________________________________
>>> pve-user mailing list
>>> pve-user@lists.proxmox.com
>>> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user
>>
>> --
>> Universität Bern
>> Phil.-nat. Fakultät
>> Institut für Informatik
>>
>> Dr. Peppo Brambilla
>> Systemadministrator
>>
>> Neubrückstrasse 10
>> 3012 Bern
>> Schweiz
>> Telefon +41 31 684 33 10
>>
>>
>> _______________________________________________
>> pve-user mailing list
>> pve-user@lists.proxmox.com
>> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user
>>
> _______________________________________________
> pve-user mailing list
> pve-user@lists.proxmox.com
> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user


-- 
Universität Bern
Phil.-nat. Fakultät
Institut für Informatik

Dr. Peppo Brambilla
Systemadministrator

Neubrückstrasse 10
3012 Bern
Schweiz
Telefon +41 31 684 33 10