[ceph-users] killing ceph-disk [was Re: ceph-volume: migration and disk partition support]

Alfredo Deza adeza at redhat.com
Tue Oct 10 04:51:11 PDT 2017


On Mon, Oct 9, 2017 at 8:50 PM, Christian Balzer <chibi at gol.com> wrote:
>
> Hello,
>
> (pet peeve alert)
> On Mon, 9 Oct 2017 15:09:29 +0000 (UTC) Sage Weil wrote:
>
>> To put this in context, the goal here is to kill ceph-disk in mimic.
>>
>> One proposal is to make it so new OSDs can *only* be deployed with LVM,
>> and old OSDs with the ceph-disk GPT partitions would be started via
>> ceph-volume support that can only start (but not deploy new) OSDs in that
>> style.
>>
>> Is the LVM-only-ness concerning to anyone?
>>
> If the provision below is met, not really.
>
>> Looking further forward, NVMe OSDs will probably be handled a bit
>> differently, as they'll eventually be using SPDK and kernel-bypass (hence,
>> no LVM).  For the time being, though, they would use LVM.
>>
> And so it begins.
> LVM does a lot of nice things, but not everything for everybody.
> It is also another layer added with all the (minor) reductions in
> performance (with normal storage, not NVMe) and of course potential bugs.
>

ceph-volume was crafted in a way that we wouldn't be forcing anyone to
a single backend (e.g. LVM). Initially it went even further,
as just being a simple orchestrator for getting devices mounted and
starting the OSD with minimal configuration and *regardless* of what
type of devices were being used.

The current status of the LVM portion is *very* robust, although it is
lacking a big chunk of feature parity with ceph-disk. I anticipate
potential bugs
anyway :)

>>
>> On Fri, 6 Oct 2017, Alfredo Deza wrote:
>> > Now that ceph-volume is part of the Luminous release, we've been able
>> > to provide filestore support for LVM-based OSDs. We are making use of
>> > LVM's powerful mechanisms to store metadata which allows the process
>> > to no longer rely on UDEV and GPT labels (unlike ceph-disk).
>> >
>> > Bluestore support should be the next step for `ceph-volume lvm`, and
>> > while that is planned we are thinking of ways to improve the current
>> > caveats (like OSDs not coming up) for clusters that have deployed OSDs
>> > with ceph-disk.
>> >
>> > --- New clusters ---
>> > The `ceph-volume lvm` deployment is straightforward (currently
>> > supported in ceph-ansible), but there isn't support for plain disks
>> > (with partitions) currently, like there is with ceph-disk.
>> >
>> > Is there a pressing interest in supporting plain disks with
>> > partitions? Or only supporting LVM-based OSDs fine?
>>
>> Perhaps the "out" here is to support a "dir" option where the user can
>> manually provision and mount an OSD on /var/lib/ceph/osd/*, with 'journal'
>> or 'block' symlinks, and ceph-volume will do the last bits that initialize
>> the filestore or bluestore OSD from there.  Then if someone has a scenario
>> that isn't captured by LVM (or whatever else we support) they can always
>> do it manually?
>>
> Basically this.
> Since all my old clusters were deployed like this, with no
> chance/intention to upgrade to GPT or even LVM.
> How would symlinks work with Bluestore, the tiny XFS bit?

In this case, we are looking to allow ceph-volume to scan currently
deployed OSDs, and get all the information
needed and save it as a plain configuration file that will be read at
boot time. That is the only other option that
is not dependent on udev/ceph-disk that doesn't mean redoing an OSD
from scratch.

It would be a one-time operation to get out of old deployment's tie
into udev/gpt/ceph-disk

>
>> > --- Existing clusters ---
>> > Migration to ceph-volume, even with plain disk support means
>> > re-creating the OSD from scratch, which would end up moving data.
>> > There is no way to make a GPT/ceph-disk OSD become a ceph-volume one
>> > without starting from scratch.
>> >
>> > A temporary workaround would be to provide a way for existing OSDs to
>> > be brought up without UDEV and ceph-disk, by creating logic in
>> > ceph-volume that could load them with systemd directly. This wouldn't
>> > make them lvm-based, nor it would mean there is direct support for
>> > them, just a temporary workaround to make them start without UDEV and
>> > ceph-disk.
>> >
>> > I'm interested in what current users might look for here,: is it fine
>> > to provide this workaround if the issues are that problematic? Or is
>> > it OK to plan a migration towards ceph-volume OSDs?
>>
>> IMO we can't require any kind of data migration in order to upgrade, which
>> means we either have to (1) keep ceph-disk around indefinitely, or (2)
>> teach ceph-volume to start existing GPT-style OSDs.  Given all of the
>> flakiness around udev, I'm partial to #2.  The big question for me is
>> whether #2 alone is sufficient, or whether ceph-volume should also know
>> how to provision new OSDs using partitions and no LVM.  Hopefully not?
>>
> I really disliked the udev/GPT stuff from the get-go and flakiness is
> being kind for sometimes completely indeterministic behavior.
>

Yep, forcing users to always fit one model seemed annoying to me. I
understand the attractiveness of the idea: just like LVM today, it
provides a narrower path for supporting more features and having a
more robust implementation.



> Since there never was an (non-disruptive) upgrade process from non-GPT
> based OSDs to GPT based ones, I wonder what changed minds here.
> Not that the GPT based users won't appreciate it.
>

We really want users to start consuming ceph-volume exclusively, but
to get there we need to find a way to deprecate ceph-disk while at the
same
time not requiring everyone to start from scratch again.

It wasn't possible to "fix" ceph-disk, and with ceph-volume we are
already doing well. My hope is that by finding the middle ground
between the two
we can eventually get to no longer support anything related to ceph-disk.

> Christian
>> sage
>> _______________________________________________
>> ceph-users mailing list
>> ceph-users at lists.ceph.com
>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
>>
>
>
> --
> Christian Balzer        Network/Systems Engineer
> chibi at gol.com           Rakuten Communications


More information about the ceph-users mailing list