From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate001.proxmox.com (gate001.proxmox.com [IPv6:2a0f:8001:1:32::40]) by lore.proxmox.com (Postfix) with ESMTPS id 9B77F1FF0B2 for ; Fri, 25 Sep 2026 14:26:56 +0200 (CEST) Received: from gate001.proxmox.com (localhost.localdomain [127.0.0.1]) by gate001.proxmox.com (Proxmox) with ESMTP id 1B0DC216B2; Fri, 25 Sep 2026 14:26:54 +0200 (CEST) From: Fiona Ebner To: pve-devel@lists.proxmox.com Subject: [PATCH qemu-server 3/7] fix #8071: volume snapshot delete: use current drive information to fix SAVC handling Date: Fri, 25 Sep 2026 14:26:27 +0200 Message-ID: <20260925122646.139215-4-f.ebner@proxmox.com> X-Mailer: git-send-email 2.47.3 In-Reply-To: <20260925122646.139215-1-f.ebner@proxmox.com> References: <20260925122646.139215-1-f.ebner@proxmox.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Bm-Milter-Handled: 55990f41-d878-4baa-be0a-ee34c49e34d2 X-Bm-Transport-Timestamp: 1790339209984 X-SPAM-LEVEL: Spam detection results: 0 AWL 0.507 Adjusted score from AWL reputation of From: address DMARC_MISSING 0.1 Missing DMARC policy KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment (newer systems) RCVD_IN_DNSWL_MED -2.3 Sender listed at https://www.dnswl.org/, medium trust SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record Message-ID-Hash: 5MDLJ7RRFPEVNXMSH5KX6573ICJEE77Y X-Message-ID-Hash: 5MDLJ7RRFPEVNXMSH5KX6573ICJEE77Y X-MailFrom: f.ebner@proxmox.com X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; loop; banned-address; emergency; member-moderation; nonmember-moderation; administrivia; implicit-dest; max-recipients; max-size; news-moderation; no-subject; digests; suspicious-header X-Mailman-Version: 3.3.10 Precedence: list List-Id: Proxmox VE development discussion List-Help: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: When removing a volume chain snapshot, the blockdev properties are re-generated in blockdev_{commit,replace,stream}() to operate upon. This must happen with the current drive information, not the one from the time the snapshot was taken. A similar bug was present before volume chain snapshots existed, and fixed by commit c6058683 ("fix #2258: select correct device when removing drive snapshot via QEMU"). This is where the handling for $attached_deviceid comes from. With snapshot-as-volume-chain, this needs to be extended to the full drive information, not just the device ID. Signed-off-by: Fiona Ebner --- src/PVE/QemuConfig.pm | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/src/PVE/QemuConfig.pm b/src/PVE/QemuConfig.pm index 3dd69ce8..39e88366 100644 --- a/src/PVE/QemuConfig.pm +++ b/src/PVE/QemuConfig.pm @@ -441,13 +441,19 @@ sub __snapshot_delete_vol_snapshot { my $running = PVE::QemuServer::Helpers::vm_running_locally($vmid); my $attached_deviceid; + # Need to operate on the currently attached device if the VM is running for QMP handling to + # work correctly. For blockdev generation, it's also important that the same drive information + # as currently attached is used, because otherwise node names and other settings might differ. if ($running) { my $conf = $class->load_config($vmid); $class->foreach_volume( $conf, sub { - my ($ds, $drive) = @_; - $attached_deviceid = "drive-$ds" if $drive->{file} eq $volid; + my ($ds, $current_drive) = @_; + if ($current_drive->{file} eq $volid) { + $attached_deviceid = "drive-$ds"; + $drive = $current_drive; + } }, ); } -- 2.47.3