So, let me explain. Doing "online storage migration" aka live storage migration is working for CEPH/NFS --> SolidFire, starting from 4.11+ Internally it is done in the same way as "virsh with --copy-storage-all --xml" in short
Longer explanation: Steps: You create new volumes on the destination storage (SolidFire in this case), set QoS etc - simply prepare the destination volumes (empty volumes atm). On source host/VM, dump VM XML, edit XML, change disk section to point to new volume path, protocol, etc - and also the IP address for the VNC (cloudstack requirement), save XMLT Then you do "virsh with --copy-storage-all --xml. myEditedVM.xml ..." stuff that does the job. Then NBD driver will be used to copy blocks from the source volumes to the destination volumes while that virsh command is working... (here's my demo, in details.. https://www.youtube.com/watch?v=Eo8BuHBnVgg&list=PLEr0fbgkyLKyiPnNzPz7XDjxnmQNxjJWT&index=5&t=2s ) This is yet to be extended/coded to support NFS-->NFS or CEPH-->CEPH or CEPH/NFS-->CEPH/NFS... should not be that much work, the logic is there (bit part of the code) Also, starting from 4.12, you can actually (I believe using identical logic) migrate only ROOT volume that are on the LOCAL storage (local disks) to another host/local storage - but DATA disks are not supported. Now...imagine the feature is there - if using CentOS7, our friends at RedHat have removed support for actually using live storage migration (unless you are paying for RHEV - but it does work fine on CentOS6, and Ubuntu 14.04+ I recall "we" had to use qemu/libvirt from the "oVirt" repo which DOES (DID) support storage live migration (normal EV packages from the Special Interest Group (2.12 tested) - did NOT include this...) I can send you step-by-step .docx guide for manually mimicking what is done (in SolidFire, but identical logic for other storages) - but not sure if that still helps you... Andrija On Wed, 25 Sep 2019 at 10:51, Melanie Desaive <m.desa...@heinlein-support.de> wrote: > Hi all, > > I am currently doing my first steps with KVM as hypervisor for > CloudStack. I was shocked to realize that currently live volume > migration between different shared storages is not supported with KVM. > This is a feature I use intensively with XenServer. > > How do you get along with this limitation? I do really expect you to > use some workarounds, or do you all only accept vm downtimes for a > storage migration? > > With my first investigation I found three techniques mentioned and > would like to ask for suggestions which to investigate deeper: > > x Eric describes a technique using snapshosts and pauses to do a live > storage migration in this mailing list tread. > x Dag suggests using virsh with --copy-storage-all --xml. > x I found articles about using virsh blockcopy for storage live > migration. > > Greetings, > > Melanie > > Am Freitag, den 02.02.2018, 15:55 +0100 schrieb Andrija Panic: > > @Dag, you might want to check with Mike Tutkowski, how he implemented > > this > > for the "online storage migration" from other storages (CEPH and NFS > > implemented so far as sources) to SolidFire. > > > > We are doing exactly the same demo/manual way (this is what Mike has > > sent > > me back in the days), so perhaps you want to see how to translate > > this into > > general things (so ANY to ANY storage migration) inside CloudStack. > > > > Cheers > > > > On 2 February 2018 at 10:28, Dag Sonstebo <dag.sonst...@shapeblue.com > > > > > wrote: > > > > > All > > > > > > I am doing a bit of R&D around this for a client at the moment. I > > > am > > > semi-successful in getting live migrations to different storage > > > pools to > > > work. The method I’m using is as follows – this does not take into > > > account > > > any efficiency optimisation around the disk transfer (which is next > > > on my > > > list). The below should answer your question Eric about moving to a > > > different location – and I am also working with your steps to see > > > where I > > > can improve the following. Keep in mind all of this is external to > > > CloudStack – although CloudStack picks up the destination KVM host > > > automatically it does not update the volume tables etc., neither > > > does it do > > > any housekeeping. > > > > > > 1) Ensure the same network bridges are up on source and destination > > > – > > > these are found with: > > > > > > [root@kvm1 ~]# virsh dumpxml 9 | grep source > > > <source file='/mnt/00e88a7b-985f-3be8-b717- > > > 0a59d8197640/d0ab5dd5- > > > e3dd-47ac-a326-5ce3d47d194d'/> > > > <source bridge='breth1-725'/> > > > <source path='/dev/pts/3'/> > > > <source path='/dev/pts/3'/> > > > > > > So from this make sure breth1-725 is up on the destionation host > > > (do it > > > the hard way or cheat and spin up a VM from same account and > > > network on > > > that host) > > > > > > 2) Find size of source disk and create stub disk in destination > > > (this part > > > can be made more efficient to speed up disk transfer – by doing > > > similar > > > things to what Eric is doing): > > > > > > [root@kvm1 ~]# qemu-img info /mnt/00e88a7b-985f-3be8-b717- > > > 0a59d8197640/d0ab5dd5-e3dd-47ac-a326-5ce3d47d194d > > > image: /mnt/00e88a7b-985f-3be8-b717-0a59d8197640/d0ab5dd5-e3dd- > > > 47ac-a326-5ce3d47d194d > > > file format: qcow2 > > > virtual size: 8.0G (8589934592 bytes) > > > disk size: 32M > > > cluster_size: 65536 > > > backing file: /mnt/00e88a7b-985f-3be8-b717-0a59d8197640/3caaf4c9- > > > eaec- > > > 11e7-800b-06b4a401075c > > > > > > ###################### > > > > > > [root@kvm3 50848ff7-c6aa-3fdd-b487-27899bf2129c]# qemu-img create > > > -f > > > qcow2 d0ab5dd5-e3dd-47ac-a326-5ce3d47d194d 8G > > > Formatting 'd0ab5dd5-e3dd-47ac-a326-5ce3d47d194d', fmt=qcow2 > > > size=8589934592 encryption=off cluster_size=65536 > > > [root@kvm3 50848ff7-c6aa-3fdd-b487-27899bf2129c]# qemu-img info > > > d0ab5dd5-e3dd-47ac-a326-5ce3d47d194d > > > image: d0ab5dd5-e3dd-47ac-a326-5ce3d47d194d > > > file format: qcow2 > > > virtual size: 8.0G (8589934592 bytes) > > > disk size: 448K > > > cluster_size: 65536 > > > > > > 3) Rewrite the new VM XML file for the destination with: > > > a) New disk location, in this case this is just a new path (Eric – > > > this > > > answers your question) > > > b) Different IP addresses for VNC – in this case 10.0.0.1 to > > > 10.0.0.2 > > > and carry out migration. > > > > > > [root@kvm1 ~]# virsh dumpxml 9 | sed -e 's/00e88a7b-985f-3be8-b717- > > > 0a59d8197640/50848ff7-c6aa-3fdd-b487-27899bf2129c/g' | sed -e 's/ > > > 10.0.0.1/10.0.0.2/g' > /root/i-2-14-VM.xml > > > > > > [root@kvm1 ~]# virsh migrate --live --persistent --copy-storage-all > > > --xml > > > /root/i-2-14-VM.xml i-2-14-VM qemu+tcp://10.0.0.2/system --verbose > > > --abort-on-error > > > Migration: [ 25 %] > > > > > > 4) Once complete delete the source file. This can be done with > > > extra > > > switches on the virsh migrate command if need be. > > > = = = > > > > > > In the simplest tests this works – destination VM remains online > > > and has > > > storage in new location – but it’s not persistent – sometimes the > > > destination VM ends up in a paused state, and I’m working on how to > > > get > > > around this. I also noted virsh migrate has a migrate- > > > setmaxdowntime which > > > I think can be useful here. > > > > > > Regards, > > > Dag Sonstebo > > > Cloud Architect > > > ShapeBlue > > > > > > On 01/02/2018, 20:30, "Andrija Panic" <andrija.pa...@gmail.com> > > > wrote: > > > > > > Actually, we have this feature (we call this internally > > > online-storage-migration) to migrate volume from CEPH/NFS to > > > SolidFire > > > (thanks to Mike Tutkowski) > > > > > > There is libvirt mechanism, where basically you start another > > > PAUSED > > > VM on > > > another host (same name and same XML file, except the storage > > > volumes > > > are > > > pointing to new storage, different paths, etc and maybe VNC > > > listening > > > address needs to be changed or so) and then you issue on > > > original > > > host/VM > > > the live migrate command with few parameters... the libvirt > > > will > > > transaprently handle the copy data process from Soruce to New > > > volumes, > > > and > > > after migration the VM will be alive (with new XML since have > > > new > > > volumes) > > > on new host, while the original VM on original host is > > > destroyed.... > > > > > > (I can send you manual for this, that is realted to SF, but > > > idea is the > > > same and you can exercies this on i.e. 2 NFS volumes on 2 > > > different > > > storages) > > > > > > This mechanism doesn't exist in ACS in general (AFAIK), except > > > for when > > > migrating to SolidFire. > > > > > > Perhaps community/DEV can help extend Mike's code to do same > > > work on > > > different storage types... > > > > > > Cheers > > > > > > > > > dag.sonst...@shapeblue.com > > > www.shapeblue.com > > > 53 Chandos Place, Covent Garden, London WC2N 4HSUK > > > @shapeblue > > > > > > > > > > > > On 19 January 2018 at 18:45, Eric Green <eric.lee.gr...@gmail.com> > > > wrote: > > > > > > > KVM is able to live migrate entire virtual machines complete > > > with > > > local > > > > volumes (see 'man virsh') but does require nbd (Network Block > > > Device) to be > > > > installed on the destination host to do so. It may need > > > installation > > > of > > > > later libvirt / qemu packages from OpenStack repositories on > > > Centos > > > 6, I'm > > > > not sure, but just works on Centos 7. In any event, I have > > > used this > > > > functionality to move virtual machines between virtualization > > > hosts > > > on my > > > > home network. It works. > > > > > > > > What is missing is the ability to live-migrate a disk from > > > one shared > > > > storage to another. The functionality built into virsh live- > > > migrates > > > the > > > > volume ***to the exact same location on the new host***, so > > > obviously is > > > > useless for migrating the disk to a new location on shared > > > storage. I > > > > looked everywhere for the ability of KVM to live migrate a > > > disk from > > > point > > > > A to point B all by itself, and found no such thing. > > > libvirt/qemu > > > has the > > > > raw capabilities needed to do this, but it is not currently > > > exposed > > > as a > > > > single API via the qemu console or virsh. It can be emulated > > > via > > > scripting > > > > however: > > > > > > > > 1. Pause virtual machine > > > > 2. Do qcow2 snapshot. > > > > 3. Detach base disk, attach qcow2 snapshot > > > > 4. unpause virtual machine > > > > 5. copy qcow2 base file to new location > > > > 6. pause virtual machine > > > > 7. detach snapshot > > > > 8. unsnapshot qcow2 snapshot at its new location. > > > > 9. attach new base at new location. > > > > 10. unpause virtual machine. > > > > > > > > Thing is, if that entire process is not built into the > > > underlying > > > > kvm/qemu/libvirt infrastructure as tested functionality with > > > a > > > defined API, > > > > there's no guarantee that it will work seamlessly and will > > > continue > > > working > > > > with the next release of the underlying infrastructure. This > > > is using > > > > multiple different tools to manipulate the qcow2 file and > > > attach/detach > > > > base disks to the running (but paused) kvm domain, and would > > > have to > > > be > > > > tested against all variations of those tools on all supported > > > Cloudstack > > > > KVM host platforms. The test matrix looks pretty grim. > > > > > > > > By contrast, the migrate-with-local-storage process is built > > > into > > > virsh > > > > and is tested by the distribution vendor and the set of tools > > > provided with > > > > the distribution is guaranteed to work with the virsh / > > > libvirt/ qemu > > > > distributed by the distribution vendor. That makes the test > > > matrix > > > for > > > > move-with-local-storage look a lot simpler -- "is this > > > functionality > > > > supported by that version of virsh on that distribution? Yes? > > > Enable > > > it. > > > > No? Don't enable it." > > > > > > > > I'd love to have live migration of disks on shared storage > > > with > > > Cloudstack > > > > KVM, but not at the expense of reliability. Shutting down a > > > virtual > > > machine > > > > in order to migrate one of its disks from one shared > > > datastore to > > > another > > > > is not ideal, but at least it's guaranteed reliable. > > > > > > > > > > > > > On Jan 19, 2018, at 04:54, Rafael Weingärtner < > > > > rafaelweingart...@gmail.com> wrote: > > > > > > > > > > Hey Marc, > > > > > It is very interesting that you are going to pick this up > > > for KVM. > > > I am > > > > > working in a related issue for XenServer [1]. > > > > > If you can confirm that KVM is able to live migrate local > > > volumes > > > to > > > > other > > > > > local storage or shared storage I could make the feature I > > > am > > > working on > > > > > available to KVM as well. > > > > > > > > > > > > > > > [1] https://issues.apache.org/jira/browse/CLOUDSTACK-10240 > > > > > > > > > > On Thu, Jan 18, 2018 at 11:35 AM, Marc-Aurèle Brothier < > > > > ma...@exoscale.ch> > > > > > wrote: > > > > > > > > > >> There's a PR waiting to be fixed about live migration with > > > local > > > volume > > > > for > > > > >> KVM. So it will come at some point. I'm the one who made > > > this PR > > > but I'm > > > > >> not using the upstream release so it's hard for me to > > > debug the > > > problem. > > > > >> You can add yourself to the PR to get notify when things > > > are > > > moving on > > > > it. > > > > >> > > > > >> https://github.com/apache/cloudstack/pull/1709 > > > > >> > > > > >> On Wed, Jan 17, 2018 at 10:56 AM, Eric Green < > > > eric.lee.gr...@gmail.com> > > > > >> wrote: > > > > >> > > > > >>> Theoretically on Centos 7 as the host KVM OS it could be > > > done > > > with a > > > > >>> couple of pauses and the snapshotting mechanism built > > > into > > > qcow2, but > > > > >> there > > > > >>> is no simple way to do it directly via virsh, the > > > libvirtd/qemu > > > control > > > > >>> program that is used to manage virtualization. It's not > > > as with > > > > issuing a > > > > >>> simple vmotion 'migrate volume' call in Vmware. > > > > >>> > > > > >>> I scripted out how it would work without that direct > > > support in > > > > >>> libvirt/virsh and after looking at all the points where > > > things > > > could go > > > > >>> wrong, honestly, I think we need to wait until there is > > > support > > > in > > > > >>> libvirt/virsh to do this. virsh clearly has the > > > capability > > > internally > > > > to > > > > >> do > > > > >>> live migration of storage, since it does this for live > > > domain > > > migration > > > > >> of > > > > >>> local storage between machines when migrating KVM domains > > > from > > > one host > > > > >> to > > > > >>> another, but that capability is not currently exposed in > > > a way > > > > Cloudstack > > > > >>> could use, at least not on Centos 7. > > > > >>> > > > > >>> > > > > >>>> On Jan 17, 2018, at 01:05, Piotr Pisz <pp...@pulab.pl> > > > wrote: > > > > >>>> > > > > >>>> Hello, > > > > >>>> > > > > >>>> Is there a chance that one day it will be possible to > > > migrate > > > volume > > > > >>> (root disk) of a live VM in KVM between storage pools (in > > > CloudStack)? > > > > >>>> Like a storage vMotion in Vmware. > > > > >>>> > > > > >>>> Best regards, > > > > >>>> Piotr > > > > >>>> > > > > >>> > > > > >>> > > > > >> > > > > > > > > > > > > > > > > > > > > -- > > > > > Rafael Weingärtner > > > > > > > > > > > > > > > > > -- > > > > > > Andrija Panić > > > > > > > > > > > > > > -- > -- > Heinlein Support GmbH > Schwedter Str. 8/9b, 10119 Berlin > > https://www.heinlein-support.de > > Tel: 030 / 40 50 51 - 62 > Fax: 030 / 40 50 51 - 19 > > Amtsgericht Berlin-Charlottenburg - HRB 93818 B > Geschäftsführer: Peer Heinlein - Sitz: Berlin > -- Andrija Panić
