qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Peter Xu <peterx@redhat.com>
Cc: Juraj Marcin <jmarcin@redhat.com>,
	qemu-devel@nongnu.org, Fabiano Rosas <farosas@suse.de>,
	Jiri Denemark <jdenemar@redhat.com>
Subject: Re: [PATCH] migration: Apply migration specific keep-alive defaults to inet socket
Date: Thu, 18 Sep 2025 16:52:29 +0100	[thread overview]
Message-ID: <aMwqvbFzGlsX23C7@redhat.com> (raw)
In-Reply-To: <aMwg-ROjbDL_z_EM@x1.local>

On Thu, Sep 18, 2025 at 11:10:49AM -0400, Peter Xu wrote:
> On Thu, Sep 18, 2025 at 03:45:21PM +0100, Daniel P. Berrangé wrote:
> > There needs to be a way to initiate post-copy recovery regardless
> > of whether we've hit a keepalive timeout. Especially if we can
> > see one QEMU in postcopy-paused, but not the other side, it
> > doesn't appear to make sense to block the recovery process.
> > 
> > The virDomainJobCancel command can do a migrate-cancel on the
> > src, but it didn't look like we could do the same on the dst.
> > Unless I've overlooked something, Libvirt needs to gain a way
> > to explicitly force both sides into the postcopy-paused state,
> > and thus be able to immediately initiate recovery.
> 
> Right, if libvirt can do that then problem should have been solved too.
> 
> > I'm fine with turning on keepalives on the socket, but IMHO the
> > out of the box behaviour should be to honour the kernel default
> > tunables unless the admin decides they want different behaviour.
> > I'm not seeing a rational for why the kernel defaults should be
> > forceably overridden in QEMU out of the box.
> 
> IMHO the rational here is that the socket here is in a special state and
> for special purpose. So we're not trying to change anything globally for
> qemu (without knowing what the socket is), but only this specific type of
> socket that is used for either precopy or postcopy live migrations.
> 
> It's special because it's always safe to have a more aggresive
> disconnection, and might be preferred versus very lengthy hangs (if
> assuming libvirt doesn't yet have way to stop the hang), especially for a
> postcopy phase.

I've already described up-thread that it is not guaranteed safe to
have aggressive disconnection. It is often safe, but there is
definitely non-negligible risk in prematurely terminating a migration
connection.

With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|



  reply	other threads:[~2025-09-18 15:53 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-09 15:01 [PATCH] migration: Apply migration specific keep-alive defaults to inet socket Juraj Marcin
2025-09-09 15:09 ` Daniel P. Berrangé
2025-09-09 21:58   ` Peter Xu
2025-09-10  7:10     ` Daniel P. Berrangé
2025-09-10 16:36       ` Peter Xu
2025-09-12 10:20         ` Daniel P. Berrangé
2025-09-12 15:02           ` Peter Xu
2025-09-15 18:23             ` Daniel P. Berrangé
2025-09-15 20:05               ` Peter Xu
2025-09-18 14:16               ` Juraj Marcin
2025-09-18 14:45                 ` Daniel P. Berrangé
2025-09-18 15:10                   ` Peter Xu
2025-09-18 15:52                     ` Daniel P. Berrangé [this message]
2025-09-19 11:59                     ` Jiří Denemark
2025-09-19 12:00                       ` Daniel P. Berrangé
2025-09-19 12:10                         ` Jiří Denemark
2025-09-19 12:04                   ` Jiří Denemark
2025-09-10 12:05   ` Juraj Marcin

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=aMwqvbFzGlsX23C7@redhat.com \
    --to=berrange@redhat.com \
    --cc=farosas@suse.de \
    --cc=jdenemar@redhat.com \
    --cc=jmarcin@redhat.com \
    --cc=peterx@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).