I jump late on the conversation because I was away from the mailing
lists last week.
We run Openstack with both nova ephemeral root disks and cinder volume
boot disks. Both are with ceph rbd backend. It is the user that flags
"boot from volume" in Horizon when starting an instance.
Everything works in both cases, and there are pros and cons as you see
from the many answers you had.
But, if I have to give you a suggestion, I would choose only 1 way to
go and stick with it.
Having all this flexibility is great for us, operators that understand
Openstack internals. But it is a nightmare for the openstack users.
First of all looking at Horizon it is very difficult to understand the
kind of root volume being used.
It is also difficult to understand that you have a snapshot of the
nova instance, and a snapshot of the cinder volume.
We have different snapshot procedures depending on the type for root
disk, and users always get confused.
When the root disk is cinder, if you snapshot from the instance page,
you will get a 0 byte glance image connected to a cinder snapshot.
A user that has a instance with a disk, should not have to understand
if the disk is managed by nova or cinder to finally backup his data
with a snapshot.
Looking at Cloud usability, I would say that mixing the two solutions
is not the best. Probably this explains the Amazon and Azure choices
you described earlier.
2017-08-01 16:50 GMT+02:00 Kimball, Conrad <conrad.kimb...@boeing.com>:
> In our process of standing up an OpenStack internal cloud we are facing the
> question of ephemeral storage vs. Cinder volumes for instance root disks.
> As I look at public clouds such as AWS and Azure, the norm is to use
> persistent volumes for the root disk. AWS started out with images booting
> onto ephemeral disk, but soon after they released Elastic Block Storage and
> ever since the clear trend has been to EBS-backed instances, and now when I
> look at their quick-start list of 33 AMIs, all of them are EBS-backed. And
> I’m not even sure one can have anything except persistent root disks in
> Azure VMs.
> Based on this and a number of other factors I think we want our user normal
> / default behavior to boot onto Cinder-backed volumes instead of onto
> ephemeral storage. But then I look at OpenStack and its design point
> appears to be booting images onto ephemeral storage, and while it is
> possible to boot an image onto a new volume this is clumsy (haven’t found a
> way to make this the default behavior) and we are experiencing performance
> problems (that admittedly we have not yet run to ground).
> So …
> · Are other operators routinely booting onto Cinder volumes instead
> of ephemeral storage?
> · What has been your experience with this; any advice?
> Conrad Kimball
> Associate Technical Fellow
> Chief Architect, Enterprise Cloud Services
> Application Infrastructure Services / Global IT Infrastructure / Information
> Technology & Data Analytics
> P.O. Box 3707, Mail Code 7M-TE
> Seattle, WA 98124-2207
> Bellevue 33-11 bldg, office 3A6-3.9
> Mobile: 425-591-7802
> OpenStack-operators mailing list
OpenStack-operators mailing list