8:21 p.m.
This is a multi-part message in MIME format.
--------------090102060102070909090704
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Hi Guys,
Thank you very much Jason, Itamar, Gadi! I'm almost there thanks to your
help.
My steps were:
- remove the old storageconnection:
$ curl -u "admin@internal:*****" -X DELETE
https://192.168.1.44:443/api/storageconnections/5636a8c3-65b6-44a4-9ba4-e...
-k
I changed version 3 -> 0 in metadata file
And did an import (export NFS):
192.168.1.44:/raid/ovirt-old/data
Currently the box is importing the VMs from the old to the new repository.
But, there's still a last challenge to overcome:
I had this VM with about 2TB storage attached on both my storage domains
(I had another one too). This poses two different problems:
- I can't import the VM from the first storage domain, since not all
disks did reside on this single storage domain. This error pops up
(which is quite logical):
Error while executing action:
downloadbak:
* Cannot import VM. VM's Image does not exist.
Yeah I know I shouldn't have split the VMs resources over multiple
storage domains... I won't make that mistake again. Is there a way to
move these disks from the second "old domain" to the first "old
domain"
so the VM can be reimported?
Furthermore, my storage machine has about 1.4TB free disk space left.
While importing, a copy of a VM is being made... and this particular one
is 2TB. So I'm going to run out of disk space while doing so; is there a
way to move instead of copy a VM while importing it?
Cheers (and thanks a lot for all the help, I really appreciate it!),
Boudewijn
--------------090102060102070909090704
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit
<html>
<head>
<meta content="text/html; charset=UTF-8"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Hi Guys,<br>
<br>
<br>
Thank you very much Jason, Itamar, Gadi! I'm almost there thanks to
your help.<br>
<br>
My steps were:<br>
<br>
- remove the old storageconnection:<br>
<br>
$ curl -u "admin@internal:*****" -X DELETE
<a class="moz-txt-link-freetext"
href="https://192.168.1.44:443/api/storageconnections/5636a8c3-65b6-44a4-9ba4-e598dc60a4e4">https://192.168.1.44:443/api/storageconnections/5636a8c3-65b6-44a4-9ba4-e598dc60a4e4</a>
-k<br>
<br>
I changed version 3 -> 0 in metadata file<br>
<br>
And did an import (export NFS):<br>
192.168.1.44:/raid/ovirt-old/data<br>
<br>
Currently the box is importing the VMs from the old to the new
repository.<br>
<br>
But, there's still a last challenge to overcome:<br>
<br>
I had this VM with about 2TB storage attached on both my storage
domains (I had another one too). This poses two different problems:<br>
<br>
- I can't import the VM from the first storage domain, since not all
disks did reside on this single storage domain. This error pops up
(which is quite logical):<br>
<br>
<div class="GJJWYTYDJYB">
<div class="GJJWYTYDEVB GJJWYTYDKYB">
<div class="gwt-HTML">Error while executing action: <br>
<br>
downloadbak:
<ul style="margin-top:0">
<li>Cannot import VM. VM's Image does not exist.</li>
</ul>
</div>
</div>
</div>
Yeah I know I shouldn't have split the VMs resources over multiple
storage domains... I won't make that mistake again. Is there a way
to move these disks from the second "old domain" to the first "old
domain" so the VM can be reimported?<br>
<br>
Furthermore, my storage machine has about 1.4TB free disk space
left. While importing, a copy of a VM is being made... and this
particular one is 2TB. So I'm going to run out of disk space while
doing so; is there a way to move instead of copy a VM while
importing it?<br>
<br>
<br>
Cheers (and thanks a lot for all the help, I really appreciate it!),<br>
<br>
Boudewijn<br>
</body>
</html>
--------------090102060102070909090704--