From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
To: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>, Fam Zheng <famz@redhat.com>,
Lai Jiangshan <laijs@cn.fujitsu.com>,
qemu block <qemu-block@nongnu.org>,
armbru@redhat.com, jcody@redhat.com,
Jiang Yunhong <yunhong.jiang@intel.com>,
Dong Eddie <eddie.dong@intel.com>,
qemu devel <qemu-devel@nongnu.org>, Max Reitz <mreitz@redhat.com>,
Gonglei <arei.gonglei@huawei.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Yang Hongyang <yanghy@cn.fujitsu.com>,
zhanghailiang <zhang.zhanghailiang@huawei.com>
Subject: Re: [Qemu-devel] [PATCH COLO v3 01/14] docs: block replication's description
Date: Fri, 8 May 2015 10:34:10 +0100 [thread overview]
Message-ID: <20150508093409.GA2126@work-vm> (raw)
In-Reply-To: <20150508084250.GA11717@stefanha-thinkpad.redhat.com>
* Stefan Hajnoczi (stefanha@redhat.com) wrote:
> On Tue, May 05, 2015 at 04:23:56PM +0100, Dr. David Alan Gilbert wrote:
> > * Stefan Hajnoczi (stefanha@redhat.com) wrote:
> > > On Fri, Apr 24, 2015 at 11:36:35AM +0200, Paolo Bonzini wrote:
> > > >
> > > >
> > > > On 24/04/2015 11:38, Wen Congyang wrote:
> > > > >> >
> > > > >> > That can be done with drive-mirror. But I think it's too early for that.
> > > > > Do you mean use drive-mirror instead of quorum?
> > > >
> > > > Only before starting up a new secondary. Basically you do a migration
> > > > with non-shared storage, and then start the secondary in colo mode.
> > > >
> > > > But it's only for the failover case. Quorum (or a new block/colo.c
> > > > driver or filter) is fine for normal colo operation.
> > >
> > > Perhaps this patch series should mirror the Secondary's disk to a Backup
> > > Secondary so that the system can be protected very quickly after
> > > failover.
> > >
> > > I think anyone serious about fault tolerance would deploy a Backup
> > > Secondary, otherwise the system cannot survive two failures unless a
> > > human administrator is lucky/fast enough to set up a new Secondary.
> >
> > I'd assumed that a higher level management layer would do the allocation
> > of a new secondary after the first failover, so no human need be involved.
>
> That doesn't help, after the first failover is too late even if it's
> done by a program. There should be no window during which the VM is
> unprotected.
>
> People who want fault tolerance care about 9s of availability. The VM
> must be protected on the new Primary as soon as the failover occurs,
> otherwise this isn't a serious fault tolerance solution.
I'm not aware of any other system that manages that, so I don't
think that's fair.
You gain a lot more availability going from a single
system to the 1+1 system that COLO (or any of the checkpointing systems)
propose, I can't say how many 9s it gets you. It's true having multiple
secondaries would get you a bit more on top of that, but you're still
a lot better off just having the one secondary.
I had thought that having >1 secondary would be a nice addition, but it's
a big change everywhere else (e.g. having to maintain multiple migration
streams, dealing with miscompares from multiple hosts).
Dave
>
> Stefan
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
next prev parent reply other threads:[~2015-05-08 9:34 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-04-03 10:01 [Qemu-devel] [PATCH COLO v3 00/14] Block replication for continuous checkpoints Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 01/14] docs: block replication's description Wen Congyang
2015-04-20 15:30 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-04-21 1:25 ` Wen Congyang
2015-04-21 15:28 ` Paolo Bonzini
2015-04-22 9:18 ` Stefan Hajnoczi
2015-04-22 9:28 ` Wen Congyang
2015-04-23 9:55 ` Stefan Hajnoczi
2015-04-23 10:11 ` Wen Congyang
2015-04-22 9:31 ` Kevin Wolf
2015-04-22 10:12 ` [Qemu-devel] " Paolo Bonzini
2015-04-23 9:00 ` Kevin Wolf
2015-04-23 9:14 ` Wen Congyang
2015-04-23 10:05 ` Paolo Bonzini
2015-04-23 10:17 ` Kevin Wolf
2015-04-23 10:33 ` Paolo Bonzini
2015-04-23 10:40 ` Kevin Wolf
2015-04-23 10:44 ` Paolo Bonzini
2015-04-23 11:35 ` Wen Congyang
2015-04-23 11:36 ` Kevin Wolf
2015-04-23 11:53 ` Paolo Bonzini
2015-04-23 12:05 ` Dr. David Alan Gilbert
2015-04-23 12:11 ` Paolo Bonzini
2015-04-23 12:19 ` Dr. David Alan Gilbert
2015-04-23 12:23 ` Paolo Bonzini
2015-04-24 2:01 ` Fam Zheng
2015-04-24 2:16 ` Wen Congyang
2015-04-24 7:47 ` Paolo Bonzini
2015-04-24 7:55 ` Wen Congyang
2015-04-24 8:58 ` Dr. David Alan Gilbert
2015-04-24 9:04 ` Paolo Bonzini
2015-04-24 9:38 ` Wen Congyang
2015-04-24 9:36 ` Paolo Bonzini
2015-04-24 9:53 ` Wen Congyang
2015-04-24 10:03 ` Paolo Bonzini
2015-04-27 9:37 ` Stefan Hajnoczi
2015-04-29 8:29 ` Paolo Bonzini
2015-04-29 8:37 ` Gonglei
2015-04-30 14:56 ` Stefan Hajnoczi
2015-05-05 15:23 ` Dr. David Alan Gilbert
2015-05-06 2:26 ` Dong, Eddie
2015-05-06 2:49 ` Fam Zheng
2015-05-08 8:42 ` Stefan Hajnoczi
2015-05-08 9:34 ` Dr. David Alan Gilbert [this message]
2015-05-08 9:39 ` Kevin Wolf
2015-05-08 9:55 ` Dr. David Alan Gilbert
2015-04-23 9:26 ` Paolo Bonzini
2015-04-23 9:37 ` Kevin Wolf
2015-04-23 9:41 ` Wen Congyang
2015-04-22 9:29 ` [Qemu-devel] [Qemu-block] " Stefan Hajnoczi
2015-04-22 9:42 ` Wen Congyang
2015-04-22 10:39 ` [Qemu-devel] " Dr. David Alan Gilbert
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 02/14] quorum: allow ignoring child errors Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 03/14] NBD client: connect to nbd server later Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 04/14] Add new block driver interfaces to control block replication Wen Congyang
2015-04-22 12:56 ` Eric Blake
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 05/14] quorum: implement block driver interfaces for " Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 06/14] NBD client: " Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 07/14] allow writing to the backing file Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 08/14] Allow creating backup jobs when opening BDS Wen Congyang
2015-04-03 11:06 ` Paolo Bonzini
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 09/14] block: Parse "backing_reference" option to reference existing BDS Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 10/14] util/hbitmap: Add an API to reset all set bits in hbitmap Wen Congyang
2015-04-03 11:05 ` Paolo Bonzini
2015-05-01 16:47 ` [Qemu-devel] [Qemu-block] " John Snow
2015-05-07 2:20 ` Wen Congyang
2015-05-07 18:32 ` John Snow
2015-05-08 0:59 ` Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 11/14] Backup: clear all bitmap when doing block checkpoint Wen Congyang
2015-04-03 11:09 ` Paolo Bonzini
2015-04-07 1:45 ` Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 12/14] qcow2: support colo Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 13/14] skip nbd_target when starting block replication Wen Congyang
2015-04-03 10:01 ` [Qemu-devel] [PATCH COLO v3 14/14] Don't allow a disk use backing reference target Wen Congyang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150508093409.GA2126@work-vm \
--to=dgilbert@redhat.com \
--cc=arei.gonglei@huawei.com \
--cc=armbru@redhat.com \
--cc=eddie.dong@intel.com \
--cc=famz@redhat.com \
--cc=jcody@redhat.com \
--cc=kwolf@redhat.com \
--cc=laijs@cn.fujitsu.com \
--cc=mreitz@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=yanghy@cn.fujitsu.com \
--cc=yunhong.jiang@intel.com \
--cc=zhang.zhanghailiang@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.