public inbox for pbs-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Hannes Laimer <h.laimer@proxmox.com>
To: pbs-devel@lists.proxmox.com
Subject: superseded: [PATCH proxmox-backup v5 0/9] fixes #6195: add support for moving groups and namespaces
Date: Tue, 31 Mar 2026 14:34:56 +0200	[thread overview]
Message-ID: <4a15ffe7-76cd-4a5b-aab0-db08b408ff69@proxmox.com> (raw)
In-Reply-To: <20260319161325.206846-1-h.laimer@proxmox.com>

superseded-by:
https://lore.proxmox.com/pbs-devel/20260331123409.198353-1-h.laimer@proxmox.com/T/#t

On 2026-03-19 17:12, Hannes Laimer wrote:
> Implements moving both backup groups and namespaces within a datastore.
> This also adds namespace locking, this allows moves to happen within a
> datastore that is otherwise in-use.
> 
> 
> # Namespace locking
> 
> To make move_group and move_namespace safe against concurrent
> operations, a namespace-level locking scheme is introduced in the first
> patch.
> 
> Locks are acquired on the directory itself at:
>   /run/proxmox-backup/locks/<store>/<ns:colon:encoded>
> 
> Two lock modes are used:
> 
> - Shared: held by operations that read from or write into a namespace
>   (backup, pull/push sync, verify per snapshot, prune per group,
>   move_group on both source and target ns).
>   Acquiring a shared lock on ns also acquires shared locks on all
>   non-root ancestors, so an exclusive lock on any ancestor blocks
>   all active operations below it.
>   Uses a 2 second timeout so that worker tasks briefly wait for a
>   concurrent move to finish rather than immediately skipping.
> 
> - Exclusive: held by operations that structurally modify a namespace
>   (move_namespace, delete_namespace). Non-blocking (timeout=0).
>   Acquiring an exclusive lock on ns also acquires shared locks on all
>   non-root ancestors, mirroring the shared variant so that two
>   concurrent structural operations on related namespaces contend
>   correctly.
> 
> Locking up the ancestor chain rather than down the subtree keeps the
> cost O(depth), bounded by MAX_NAMESPACE_DEPTH (8).
> 
> Verify and prune skip gracefully when a namespace lock cannot be
> acquired, since a concurrent move is a transient condition.
> create_locked_backup_group now returns (owner, ns_guard, group_guard),
> all callers updated.
> 
> # Moving
> 
> ## Groups
>   1. lock source ns (shared), lock target ns (shared), lock group (exclusive)
>   2. create target type directory
>   3. FS: rename group directory
>      S3: copy objects, rename cache, delete source objects
> 
> ## Namespace
>   1. lock source ns (exclusive), lock target ns (exclusive)
>   2. FS: rename namespace directory
>      S3: - create target ns dirs and markers, both on S3 and local cache
>          - move groups one by one, if copying fails the group stays at
>            the source and can be moved afterwards manually
>          - clean up empty source ns dirs
> 
> 
> v5, thanks @Chris!:
>  - lock dir instead of `.ns-lock` file
>  - explicitly drop ns lock guards in specific order
>  - improve cleanup of partially failed s3 moves, we now create the local
>    empty dir+owner file before we start copying s3 objects, if any of
>    the s3 ops fail, the dir stays behind and can be deleted through the
>    UI(which also triggers a prefix cleanup on the s3 storage)
>  - update parameters for `DataStore::lookup_datastore()`
>  - ui: re-ordered actions, `move` now next to `verify`
>  - ui: add move to right-click context menu
>  - ui: show empty groups in the UI
>  - add cli commands for both ns and group moves
>  - add 2s ns lock timeout for worker tasks
> 
> *note*: given the UI change to show empty groups it could make sense to
> not auto-delete a group if the last snapshot is deleted. For this series
> though that is not relevant since we just need empty groups to be
> deletable through the UI for partially failed s3 moves
> 
> 
> 
> Hannes Laimer (9):
>   ui: show empty groups
>   datastore: add namespace-level locking
>   datastore: add move_group
>   datastore: add move_namespace
>   api: add PUT endpoint for move_group
>   api: add PUT endpoint for move_namespace
>   ui: add move group action
>   ui: add move namespace action
>   cli: add move-namespace and move-group commands
> 
>  pbs-datastore/src/backup_info.rs            | 261 ++++++++++++++-
>  pbs-datastore/src/datastore.rs              | 345 +++++++++++++++++++-
>  src/api2/admin/datastore.rs                 |  78 ++++-
>  src/api2/admin/namespace.rs                 |  86 ++++-
>  src/api2/backup/environment.rs              |   4 +
>  src/api2/backup/mod.rs                      |  14 +-
>  src/api2/tape/restore.rs                    |   9 +-
>  src/backup/verify.rs                        |  19 +-
>  src/bin/proxmox_backup_manager/datastore.rs |  85 ++++-
>  src/server/prune_job.rs                     |  11 +
>  src/server/pull.rs                          |   8 +-
>  src/server/push.rs                          |   6 +
>  www/Makefile                                |   2 +
>  www/datastore/Content.js                    | 134 +++++++-
>  www/form/NamespaceSelector.js               |  11 +
>  www/window/GroupMove.js                     |  56 ++++
>  www/window/NamespaceMove.js                 |  79 +++++
>  17 files changed, 1166 insertions(+), 42 deletions(-)
>  create mode 100644 www/window/GroupMove.js
>  create mode 100644 www/window/NamespaceMove.js
> 





      parent reply	other threads:[~2026-03-31 12:34 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-19 16:13 Hannes Laimer
2026-03-19 16:13 ` [PATCH proxmox-backup v5 1/9] ui: show empty groups Hannes Laimer
2026-03-24 10:28   ` Dominik Csapak
2026-03-19 16:13 ` [PATCH proxmox-backup v5 2/9] datastore: add namespace-level locking Hannes Laimer
2026-03-24 13:00   ` Fabian Grünbichler
2026-03-19 16:13 ` [PATCH proxmox-backup v5 3/9] datastore: add move_group Hannes Laimer
2026-03-24 13:00   ` Fabian Grünbichler
2026-03-25  6:20     ` Hannes Laimer
2026-03-19 16:13 ` [PATCH proxmox-backup v5 4/9] datastore: add move_namespace Hannes Laimer
2026-03-24 13:00   ` Fabian Grünbichler
2026-03-25  6:32     ` Hannes Laimer
2026-03-25  8:55       ` Fabian Grünbichler
2026-03-19 16:13 ` [PATCH proxmox-backup v5 5/9] api: add PUT endpoint for move_group Hannes Laimer
2026-03-19 16:13 ` [PATCH proxmox-backup v5 6/9] api: add PUT endpoint for move_namespace Hannes Laimer
2026-03-19 16:13 ` [PATCH proxmox-backup v5 7/9] ui: add move group action Hannes Laimer
2026-03-24 10:41   ` Dominik Csapak
2026-03-19 16:13 ` [PATCH proxmox-backup v5 8/9] ui: add move namespace action Hannes Laimer
2026-03-24 10:56   ` Dominik Csapak
2026-03-31 12:27     ` Hannes Laimer
2026-03-31 12:29     ` Hannes Laimer
2026-03-19 16:13 ` [PATCH proxmox-backup v5 9/9] cli: add move-namespace and move-group commands Hannes Laimer
2026-03-31 12:34 ` Hannes Laimer [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=4a15ffe7-76cd-4a5b-aab0-db08b408ff69@proxmox.com \
    --to=h.laimer@proxmox.com \
    --cc=pbs-devel@lists.proxmox.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
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal