qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Fabiano Rosas <farosas@suse.de>
To: Kevin Wolf <kwolf@redhat.com>, qemu-block@nongnu.org
Cc: kwolf@redhat.com, hreitz@redhat.com, stefanha@redhat.com,
	pkrempa@redhat.com, peterx@redhat.com, qemu-devel@nongnu.org
Subject: Re: [PATCH v2 00/15] block: Managing inactive nodes (QSD migration)
Date: Fri, 31 Jan 2025 10:53:36 -0300	[thread overview]
Message-ID: <877c6bgkdb.fsf@suse.de> (raw)
In-Reply-To: <20250130171240.286878-1-kwolf@redhat.com>

Kevin Wolf <kwolf@redhat.com> writes:

> This series adds a mechanism that allows the user or management tool to
> manually activate and inactivate block nodes instead of fully relying on
> the automatic management in the migration code.
>
> One case where this is needed is for migration with shared storage and
> devices backed by qemu-storage-daemon, which as an external process is
> not involved in the VM migration. Management tools can manually
> orchestrate the handover in this scenario. The new qemu-iotests case
> qsd-migrate demonstrates this.
>
> There are other cases without qemu-storage-daemon where manual
> management is necessary. For example, after migration, the destination
> VM only activates images on 'cont', but after migrating a paused VM, the
> user may want to perform operations on a block node while the VM is
> still paused.
>
> This series adds support for block exports on an inactive node (needed
> for shared storage migration with qemu-storage-daemon) only to NBD.
> Adding it to other export types will be done in a future series.
>
> v2:
> - Added a comprehensive test case that tests how inactive nodes
>   interoperate with many operations
> - Added a couple of fixes for bugs uncovered by the tests (that would
>   usually lead to crashes when an unsupported operation is performed on
>   inactive nodes)
> - Added 'active' status to query-block information
>
> Kevin Wolf (15):
>   block: Add 'active' field to BlockDeviceInfo
>   block: Inactivate external snapshot overlays when necessary
>   migration/block-active: Remove global active flag
>   block: Don't attach inactive child to active node
>   block: Allow inactivating already inactive nodes
>   block: Fix crash on block_resize on inactive node
>   block: Add option to create inactive nodes
>   block: Add blockdev-set-active QMP command
>   block: Support inactive nodes in blk_insert_bs()
>   block/export: Don't ignore image activation error in blk_exp_add()
>   block/export: Add option to allow export of inactive nodes
>   nbd/server: Support inactive nodes
>   iotests: Add filter_qtest()
>   iotests: Add qsd-migrate case
>   iotests: Add (NBD-based) tests for inactive nodes
>
>  qapi/block-core.json                          |  44 ++-
>  qapi/block-export.json                        |  10 +-
>  include/block/block-common.h                  |   1 +
>  include/block/block-global-state.h            |   6 +
>  include/block/export.h                        |   3 +
>  migration/migration.h                         |   3 -
>  block.c                                       |  62 +++-
>  block/block-backend.c                         |  16 +-
>  block/export/export.c                         |  29 +-
>  block/monitor/block-hmp-cmds.c                |   5 +-
>  block/qapi.c                                  |   1 +
>  blockdev.c                                    |  48 +++
>  migration/block-active.c                      |  46 ---
>  migration/migration.c                         |   8 -
>  nbd/server.c                                  |  17 +
>  tests/qemu-iotests/iotests.py                 |   8 +
>  tests/qemu-iotests/041                        |   4 +-
>  tests/qemu-iotests/165                        |   4 +-
>  tests/qemu-iotests/184.out                    |   2 +
>  tests/qemu-iotests/191.out                    |  16 +
>  tests/qemu-iotests/273.out                    |   5 +
>  tests/qemu-iotests/tests/copy-before-write    |   3 +-
>  tests/qemu-iotests/tests/inactive-node-nbd    | 303 ++++++++++++++++++
>  .../qemu-iotests/tests/inactive-node-nbd.out  | 239 ++++++++++++++
>  tests/qemu-iotests/tests/migrate-bitmaps-test |   7 +-
>  tests/qemu-iotests/tests/qsd-migrate          | 132 ++++++++
>  tests/qemu-iotests/tests/qsd-migrate.out      |  51 +++
>  27 files changed, 986 insertions(+), 87 deletions(-)
>  create mode 100755 tests/qemu-iotests/tests/inactive-node-nbd
>  create mode 100644 tests/qemu-iotests/tests/inactive-node-nbd.out
>  create mode 100755 tests/qemu-iotests/tests/qsd-migrate
>  create mode 100644 tests/qemu-iotests/tests/qsd-migrate.out

Series:

Acked-by: Fabiano Rosas <farosas@suse.de>

I checked that this series doesn't regress the original double
inactivate issue fixed by Peter Xu [1]. I also ported the tests[2] on
top of this and everything looks good. Thanks!

1- 8597af7615 (migration/block: Rewrite disk activation, 2024-12-06)
2- https://lore.kernel.org/r/20241125144612.16194-6-farosas@suse.de



      parent reply	other threads:[~2025-01-31 13:54 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-30 17:12 [PATCH v2 00/15] block: Managing inactive nodes (QSD migration) Kevin Wolf
2025-01-30 17:12 ` [PATCH v2 01/15] block: Add 'active' field to BlockDeviceInfo Kevin Wolf
2025-01-30 19:30   ` Eric Blake
2025-02-03 18:47   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 02/15] block: Inactivate external snapshot overlays when necessary Kevin Wolf
2025-01-30 19:46   ` Eric Blake
2025-02-03 18:48   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 03/15] migration/block-active: Remove global active flag Kevin Wolf
2025-01-30 19:50   ` Eric Blake
2025-02-04 15:50     ` Kevin Wolf
2025-02-03 18:49   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 04/15] block: Don't attach inactive child to active node Kevin Wolf
2025-01-30 20:08   ` Eric Blake
2025-02-03 18:50   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 05/15] block: Allow inactivating already inactive nodes Kevin Wolf
2025-01-30 20:09   ` Eric Blake
2025-02-03 18:51   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 06/15] block: Fix crash on block_resize on inactive node Kevin Wolf
2025-01-30 20:11   ` Eric Blake
2025-02-03 18:52   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 07/15] block: Add option to create inactive nodes Kevin Wolf
2025-01-30 20:17   ` Eric Blake
2025-02-03 18:53   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 08/15] block: Add blockdev-set-active QMP command Kevin Wolf
2025-01-30 20:22   ` Eric Blake
2025-02-03 18:54   ` Stefan Hajnoczi
2025-01-30 17:12 ` [PATCH v2 09/15] block: Support inactive nodes in blk_insert_bs() Kevin Wolf
2025-01-30 20:26   ` Eric Blake
2025-02-03 18:55   ` Stefan Hajnoczi
2025-01-31  9:50 ` [PATCH v2 10/15] block/export: Don't ignore image activation error in blk_exp_add() Kevin Wolf
2025-02-03 16:30   ` Eric Blake
2025-02-03 18:58   ` Stefan Hajnoczi
2025-02-04 15:54     ` Kevin Wolf
2025-01-31  9:50 ` [PATCH v2 11/15] block/export: Add option to allow export of inactive nodes Kevin Wolf
2025-01-31 13:41   ` Fabiano Rosas
2025-02-04 15:44     ` Kevin Wolf
2025-02-03 19:12   ` Eric Blake
2025-02-03 19:17   ` Stefan Hajnoczi
2025-01-31  9:50 ` [PATCH v2 12/15] nbd/server: Support " Kevin Wolf
2025-02-03 19:17   ` Eric Blake
2025-02-03 19:19   ` Stefan Hajnoczi
2025-02-04 17:10     ` Kevin Wolf
2025-01-31  9:50 ` [PATCH v2 13/15] iotests: Add filter_qtest() Kevin Wolf
2025-02-03 19:19   ` Eric Blake
2025-01-31  9:50 ` [PATCH v2 14/15] iotests: Add qsd-migrate case Kevin Wolf
2025-02-03 19:35   ` Eric Blake
2025-02-03 21:09     ` Kevin Wolf
2025-01-31  9:50 ` [PATCH v2 15/15] iotests: Add (NBD-based) tests for inactive nodes Kevin Wolf
2025-02-03 19:49   ` Eric Blake
2025-02-04 16:30     ` Kevin Wolf
2025-01-31 13:53 ` Fabiano Rosas [this message]

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=877c6bgkdb.fsf@suse.de \
    --to=farosas@suse.de \
    --cc=hreitz@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=peterx@redhat.com \
    --cc=pkrempa@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.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 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).