On Wed, 23 Apr 2014, Steve Dainard wrote:
> I have other VM's with the same amount of snapshots without this problem.
> No conclusion jumping going on. More interested in what the best practice
> is for VM's that accumulate snapshots over time.
For some real world context, we seem to accumulate snapshots
using our local approach, and are not that focused on, or
attentive about removing them. The 'highwater mark' of 39, on
a machine that has been around since it was provisioned:
2010-01-05
[root@xxx backups]# ./count-snapshots.sh | sort -n | tail -3
38 vm_64099
38 vm_98036
39 vm_06359
Accumulating large numbers of snapshots seems more the
function of pets, than ephemeral 'cattle'
I wrote the first paragraph without looking up the 'owners' of
the images. As I dereference the VM id's, all of the top ten
in that list turn out to be mailservers, radius servers, name
servers, and such, where the business unit owners chose not
(or neglect) to 'winnow' their herd. There are no ephemeral
use units in the top ten
-- Russ herrold
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
please note there is a recommended limit of having no more than 500
snapshots per block storage domain due to some LVM performance issues
with high number of LVs. each disk/snapshot is an LV.
NFS doesn't have this limitation.