[ovirt-users] Storage over allocation ratio value larger than 125% will pause vms

[Grundmann, Christian]

Hi,
I have the same problem with thin provisioned VMs (or VMs with snapshot)
It got a lot better (but not fixed) using this guide https://access.redhat.com/solutions/130843 to adjust the tunables

Maybe it helps
Christian

Von: users-bounces at ovirt.org [mailto:users-bounces at ovirt.org] Im Auftrag von Yaniv Kaul
Gesendet: Montag, 07. Dezember 2015 15:49
An: Jan Bernard Wevers <jan.bernard.wevers at group2000.com>
Cc: users at ovirt.org
Betreff: Re: [ovirt-users] Storage over allocation ratio value larger than 125% will pause vms

On Mon, Dec 7, 2015 at 12:38 AM, Jan Bernard Wevers <jan.bernard.wevers at group2000.com<mailto:jan.bernard.wevers at group2000.com>> wrote:
I’m using ovirt 3.6 and an iscsi storage domain. Almost all vms are using
thin provisioning with a virtual size of 40GB and a typical physical size of
8 GB.
Now I’m facing the fact that ovirt is pausing my vms due to the following
event message:
(Example):  “VM vm_01_16_centos6 has been paused due to no Storage space
error.”

oVirt is not pausing VMs. QEMU/KVM does, when it gets a storage error (in order to prevent possible corruption).
Can you please attach VDSM logs from the relevant host where this is happening?
TIA,
Y.


When I look at the iscsi storage domain the over allocation ratio value is
about 127%, but of the 2799GB there is still more than 1500GB available.
My questions:
- Why ovirt is pausing the vms?
- Can I enlarge the over allocation ratio before ovirt is complaining about
storage space?


_______________________________________________
Users mailing list
Users at ovirt.org<mailto:Users at ovirt.org>
http://lists.ovirt.org/mailman/listinfo/users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/users/attachments/20151207/4b61a75e/attachment-0001.html>