suffixes stay supported for backwards-compatibility
for API/CLI users as well as parsing backup configurations. The
function is currently also used for parsing the 'target-size' Ceph
pool option (still restricted via schema to single-letter suffixes).
Signed-off-by: Fiona Ebner
---
Cha
(or otherwise) backwards-migrations of guests.
Changes in v2:
* Only add parsing support for now.
common:
Fiona Ebner (1):
schema: support sizes with verbose suffixes {K,M,G,T}iB
src/PVE/JSONSchema.pm | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
manager:
Fiona Ebner (1
The single-letter suffixes are ambiguous and especially in the context
of disks, the powers of ten are usually used. Use exact suffixes
instead when logging the archive file size.
Signed-off-by: Fiona Ebner
---
No changes in v2.
PVE/VZDump.pm | 8
1 file changed, 4 insertions(+), 4
Am 01.06.25 um 11:51 schrieb Thomas Lamprecht:
> Am 12.05.25 um 15:00 schrieb Fiona Ebner:
>> Am 12.05.25 um 14:41 schrieb Fiona Ebner:
>>> Maybe best is to wait for PVE 9 with this and do a parse+write for all
>>> guest configs (including their snapshots) in the pve
Suggested-by: Thomas Lamprecht
Signed-off-by: Fiona Ebner
---
Changes in v2:
* Instead of removing the warning, provide more context.
block/gluster.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/block/gluster.c b/block/gluster.c
index c3a991..71c4a60b02 100644
Am 02.06.25 um 08:33 schrieb Thomas Lamprecht:
> Am 11.04.25 um 11:37 schrieb Fiona Ebner:
>> Proxmox VE will not drop the GlusterFS plugin during the PVE 8.x
>> support lifecycle. Avoid scaring users early.
>
> Those warnings from QEMU come up as standard info in the task lo
n
qemu_savevm_state_complete*")
Also squash some related patches.
Fiona Ebner (2):
update submodule and patches to QEMU 10.0.2
squash some related patches
...d-support-for-sync-bitmap-mode-never.patch | 30 +-
...-support-for-conditional-and-always-.patch | 2 +-
...check-for-bitma
Applied, thanks! I changed the path/pool to not re-use existing ones for
different storages, just for future-proofing/more realism.
[1/1] tests: qemu_img_convert: add rbd and btrfs snapshots convert test
commit: 1fd1ca60f94c68702a6ebda1000e0e9d5fc7f50f
_
Am 28.05.25 um 11:08 schrieb DERUMIER, Alexandre via pve-devel:
> perl question: how to call a resursive private sub ? (it don't seem to
> allow it)
AFAIK, you can do it by declaring it up-front:
> [I] root@pve8a1 ~# cat asdf.pm
> use strict;
> use warnings;
>
> package Private;
>
> my sub add
Please use present tense in the commit title/subject like is customary
in Proxmox VE development.
Am 27.05.25 um 14:39 schrieb Alexander Abraham:
> The section about "Creating SWAP partitions" lives in its own
> file and the warning about creating SWAP partitions on ZFS was
> moved back to the ori
As reported in the community forum [0], it was not possible to use the
extended schema that supports disk allocation with 'import-from' for
'qm import'. Align it more closely to the VM creation endpoint.
[0]: https://forum.proxmox.com/threads/166734/
Signed-off-by: Fion
/dev'
explicitly for setups that require it.
Signed-off-by: Fiona Ebner
---
Changes in v2:
* Cleaner/more robust approach, add storage configuration setting.
src/PVE/Storage/LunCmd/Comstar.pm | 3 ++-
src/PVE/Storage/LunCmd/Iet.pm | 5 ++--
src/PVE/Storage/LunCmd/Istgt.pm | 5
This reduces the potential breakage from commit "fix #5071: zfs over
iscsi: add 'zfs-base-path' configuration option". Only setups where
'/dev/zvol' exists, but is not a valid base, will still be affected.
Signed-off-by: Fiona Ebner
---
This one is optional, not s
Am 27.05.25 um 12:59 schrieb Fiona Ebner:
> sub get_base {
> -return '/dev';
> +if (!$base_dir) {
> + $base_dir = -e '/dev/zvol' ? '/dev/zvol' : '/dev';
> +}
> +return $base_dir;
> }
>
Of course, this is co
Am 27.05.25 um 12:35 schrieb Alexander Abraham:
> The section about "Creating SWAP partitions" lives in its own
> file and the warning about creating SWAP partitions on ZFS was
> moved back to the original file with all documentation about
> working with ZVols.
>
> Signed-off-by: Alexander Abraham
l' does not exist
for backwards compatibility.
Signed-off-by: Fiona Ebner
---
If other exotic setups pop up where this is not enough, it can still
be made into an actual storage configuration property. From the
existing reports, this should be enough however and the logic makes
sense as the def
Am 27.05.25 um 11:34 schrieb Michael Köppl:
> On 5/20/25 16:03, Fiona Ebner wrote:
>>> @@ -1558,13 +1564,17 @@ sub vmconfig_apply_pending {
>>> next if $selection && !$selection->{$opt};
>>> eval {
>>> my $mp = $class->parse_
On Fri, 18 Apr 2025 10:36:58 +0200, Fabian Grünbichler wrote:
> prevent a warning about $ver being an uninitialized value in case the tunnel
> didn't provide a version string at all, for example because it already closed
> cause of a prior error.
Applied, thanks!
[1/1] tunnel: improve version err
Am 14.06.24 um 13:29 schrieb Fiona Ebner:
> While archives with unknown or undetermined subtype could be shown,
> this is only for autocompletion, so users can still specify those
> manually if required.
>
> Signed-off-by: Fiona Ebner
> ---
> src/PVE/LXC.pm | 1 +
> 1 f
Am 09.02.24 um 14:08 schrieb Fiona Ebner:
> Nowadays, TFA is configured in the respective window, so only legacy
> values require some special treatment. The only real use case for
> actual legacy keys is syncing them from LDAP. I wasn't sure if it
> still makes sense to keep t
Am 22.05.25 um 15:53 schrieb Alexandre Derumier via pve-devel:
> This is part of my work on qcow2 external snapshot, but could improve current
> qcow2 linked clone
>
> This patch serie move qemu_img_create to common helpers,
> and enable preallocation on backed_image to increase performance
>
>
Yes, this is a better place for such helpers :)
Am 22.05.25 um 15:53 schrieb Alexandre Derumier via pve-devel:
> Signed-off-by: Alexandre Derumier
> ---
> src/PVE/Storage/Common.pm | 53 ++
> src/PVE/Storage/GlusterfsPlugin.pm | 2 +-
> src/PVE/Storage/Plugi
Am 22.05.25 um 15:53 schrieb Alexandre Derumier via pve-devel:
>
> and use it for plugin linked clone
>
> This also enable extended_l2=on, as it's mandatory for backing file
> preallocation.
>
> Preallocation was missing previously, so it should increase performance
> for linked clone now (aroun
superseded by:
https://lore.proxmox.com/pve-devel/20250526141941.1654201-1-f.eb...@proxmox.com/T/
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
luate to false in PVE/AccessControl.pm's parse_user_config():
> if (!verify_poolname($pool, 1)) {
> warn "user config - ignore pool '$pool' - invalid characters in pool
> name\n";
> next;
> }
It's likely that it would cause other issues
Am 26.05.25 um 12:38 schrieb Thomas Lamprecht:
> Am 26.05.25 um 10:45 schrieb Fiona Ebner:
>> Same rationale as 4fd2027e ("service: add restart on-failure to
>> pveproxy and pvedaemon") which added the setting for the pveproxy and
>> pvedaemon services.
>
Same rationale as 4fd2027e ("service: add restart on-failure to
pveproxy and pvedaemon") which added the setting for the pveproxy and
pvedaemon services.
Suggested for pvestatd in the community forum:
https://forum.proxmox.com/threads/165597/post-773210
Signed-off-by: Fiona Ebner
---
.
Signed-off-by: Fiona Ebner
---
New in v2.
src/PVE/Storage/ISCSIDirectPlugin.pm | 3 +++
src/PVE/Storage/Plugin.pm| 13 -
src/PVE/Storage/RBDPlugin.pm | 2 ++
src/PVE/Storage/ZFSPlugin.pm | 3 +++
src/PVE/Storage/ZFSPoolPlugin.pm | 2 ++
5 files changed
be usable for virtual machines anymore.
Therefore, this is intended for the next major release, Proxmox VE 9.
Signed-off-by: Fiona Ebner
---
New in v2.
ApiChangeLog | 13 +
src/PVE/Storage.pm | 4 ++--
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/ApiC
uture.
Co-developed-by: Alexandre Derumier
Signed-off-by: Fiona Ebner
---
Changes in v2:
* Remove $snapname, is added in later patch as an option.
* Adapt to updated QEMU API.
src/PVE/Storage/RBDPlugin.pm | 42
1 file changed, 42 insertions(+)
diff --git
irectsync │ off │ on │ off│
├─┼─┼──┼┤
│unsafe │ on │ off │ on │
└─┴─┴──┴────────┘
Table from 'man kvm'.
Signed-off-by: Fiona Ebner
s-qcow2 exported via NBD or FUSE or external backup
provider restore providing an NBD export for the provider to write to.
QEMU patches submitted upstream here:
https://lore.kernel.org/qemu-devel/20250515112908.383693-1-f.eb...@proxmox.com/
qemu:
Fiona Ebner (2):
block/rbd: support selected k
ation option is most convenient.
Signed-off-by: Fiona Ebner
---
New in v2.
block/rbd.c | 8
qapi/block-core.json | 5 -
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/block/rbd.c b/block/rbd.c
index 3928d8fee4..0eac4c1d3e 100644
--- a/block/rbd.c
+++ b/b
' to 'writeback' yields a substantial improvement
there [0].
The function qemu_rbd_extract_key_value_pairs() was copied/adapted
from the existing qemu_rbd_extract_encryption_create_options().
[0]: https://bugzilla.proxmox.com/show_bug.cgi?id=3329#c9
Signed-off-by: Fiona Ebner
Signed-off-by: Fiona Ebner
---
Changes in v2:
* Remove $snapname, is added in later patch as an option.
src/PVE/Storage/ISCSIDirectPlugin.pm | 14 ++
1 file changed, 14 insertions(+)
diff --git a/src/PVE/Storage/ISCSIDirectPlugin.pm
b/src/PVE/Storage/ISCSIDirectPlugin.pm
index
9 being a major release and most popular
plugins not using special protocols like 'rbd://', this seems
acceptable.
For NBD, etc. qemu-server should construct the blockdev object.
Signed-off-by: Fiona Ebner
---
Changes in v2:
* Remove $snapname, is added in later patch as an option.
* C
ZFS does not have a filesystem_path() method, so the default
implementation for qemu_blockdev_options() cannot be re-used. This is
most likely, because snapshots are currently not directly accessible
via a filesystem path in the Proxmox VE storage layer.
Signed-off-by: Fiona Ebner
---
New in v2
Reported-by: Alexandre Derumier
Signed-off-by: Fiona Ebner
---
New in v2
src/PVE/Storage/ZFSPlugin.pm | 16
1 file changed, 16 insertions(+)
diff --git a/src/PVE/Storage/ZFSPlugin.pm b/src/PVE/Storage/ZFSPlugin.pm
index 94cb11f..89eff3c 100644
--- a/src/PVE/Storage
Am 23.05.25 um 10:30 schrieb DERUMIER, Alexandre:
>>>
>>> if ($path =~ m|^/|) {
>>> # The 'file' driver only works for regular files. The check
>>> below is taken from
>>> # block/file-posix.c:hdev_probe_device() in QEMU. Do not
> bother >>with detecting 'host_cdrom'
>>> # devices h
Am 23.05.25 um 11:34 schrieb DERUMIER, Alexandre:
>>> I intentionally do not handle CD-ROMs, qemu-server should be
>>> concerned
>>> with doing that. There is a comment about this ;)
>
> I mean, could it better to have something like this ? :
>
>
> sub qemu_blockdev_options {
> my ($cfg, $vo
Am 23.05.25 um 10:50 schrieb DERUMIER, Alexandre:
>>
>> Also, for lvm volume, they are not currently activate at vm command
>> line generation. (but anyway, I'll need it to retrieve backing chain,
>> so maybe it's not a problem)
>
>>> Thanks for testing! I'll add an activate_volume() call in the
>
Am 23.05.25 um 11:15 schrieb DERUMIER, Alexandre:
>>> Good point! I guess it's better to also do the activation in qemu-
>>> server
>>> then, so that we can match it up nicely without going over package
>>> boundaries.
>
> Ok, I'll add the activate in qemu-server too
>
>
>
> Just found another
Am 23.05.25 um 10:42 schrieb DERUMIER, Alexandre:
>>> no, the plan is to drop the storage plugin for it with Proxmox VE 9.
>>> We'll warn about it in the pve8to9 upgrade script. QEMU is expected
>>> to
>>> drop support with 10.1 in a few months. People that want to continue
>>> using it, can still
Hi Alexandre,
Am 23.05.25 um 10:19 schrieb DERUMIER, Alexandre:
> do we still support glusterfs for pve9 (as it's deprecated)?
no, the plan is to drop the storage plugin for it with Proxmox VE 9.
We'll warn about it in the pve8to9 upgrade script. QEMU is expected to
drop support with 10.1 in a fe
Am 14.05.25 um 11:36 schrieb Fiona Ebner:
> As reported in the enterprise support, the usage percentage presented
> by Proxmox VE can be quite different from what Ceph itself shows when
> compression is used on the pool. The reason is that Proxmox VE used
> the 'stored' va
would show the default value qcow2 in a disabled
> state, which could confuse users.
>
> Signed-off-by: Markus Frank
Reviewed-by: Fiona Ebner
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
On Mon, 19 May 2025 16:43:57 +0200, Dominik Csapak wrote:
> If no vfio device is present during migration, and the transferred
> (main) memory did not change between loop cycles, we get a warning:
>
> Use of uninitialized value $last_vfio_transferred in string ne
>
> To silence that, check if
Am 07.03.25 um 10:25 schrieb Markus Frank:
> Prerequisite for "ui: restore window: add diskformat option"
>
> The hide condition is copied from the format selector item in the same
> file.
>
> Signed-off-by: Markus Frank
With the eslint issue below fixed:
Reviewed-
ned-off-by: Markus Frank
Looks good to me, but some slight changes would be nice, see below:
Reviewed-by: Fiona Ebner
> ---
> v3:
> * added requires => 'archive' to disk-format
> * changed descriptions
> * added missing imports/uses
> * check if disk_format i
Am 07.03.25 um 10:25 schrieb Markus Frank:
> Signed-off-by: Markus Frank
> ---
> v3:
> * removed 'optional => 1'
> * changed description
>
> src/PVE/Storage/Plugin.pm | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/src/PVE/Storage/Plugin.pm b/src/PVE/Storage/Plugin.pm
> index 65c
Am 30.04.25 um 12:45 schrieb Maximiliano Sandoval:
> Signed-off-by: Maximiliano Sandoval
> ---
> www/manager6/window/Restore.js | 10 ++
> 1 file changed, 10 insertions(+)
>
> diff --git a/www/manager6/window/Restore.js b/www/manager6/window/Restore.js
> index 22900868..8d6dfabe 100644
>
's likely that it would cause other issues as well if properly
handled there, so prohibit it in the schema.
Signed-off-by: Fiona Ebner
---
src/PVE/AccessControl.pm | 5 +
1 file changed, 5 insertions(+)
diff --git a/src/PVE/AccessControl.pm b/src/PVE/AccessControl.pm
index 1c79656..224ae
Am 20.05.25 um 11:08 schrieb Michael Köppl:
> If the storage backing the volume was removed, the volume will be marked
> as unused when pending actions are processed (e.g. volume was detached).
> Previously, the volume would simply disappear from the config.
Why? If the storage doesn't exist, the
; Co-authored-by: Stefan Hrdlicka
As already, mentioned, needs to be adapted.
> Signed-off-by: Michael Köppl
With the comments addressed
Reviewed-by: Fiona Ebner
> ---
> PVE/QemuServer.pm | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/PVE
Am 20.05.25 um 11:08 schrieb Michael Köppl:
> Align error handling behavior when checking for linked clones with the
> rest of destroy_vm's error handling approach. Unlike the LXC destroy
> implementation, this version continues execution when encountering
> errors during the check, since:
>
> 1.
uction of the container template will
> now fail, informing the user that the base volume is still in use by
> the linked clone.
>
> Originally-by: Stefan Hrdlicka
> Signed-off-by: Michael Köppl
With the comments addressed, consider this:
Reviewed-by: Fiona Ebner
>
Am 20.05.25 um 11:08 schrieb Michael Köppl:
> Detaching a mount point with a removed underlying storage causes it to
> be labeled as an unused disk. With this change, removing an unused disk
> with a removed underlying storage causes it to be removed from the
> configuration instead of failing. Del
Am 20.05.25 um 15:45 schrieb Fiona Ebner:
> Am 20.05.25 um 11:08 schrieb Michael Köppl:
>> The volume ID returned by parse_volume() is used here as opposed to
>> $conf->{$opt} used for regular mount points because $conf->{$opt} is not
>> necessarily a valid volume ID in
Am 20.05.25 um 11:08 schrieb Michael Köppl:
> The volume ID returned by parse_volume() is used here as opposed to
> $conf->{$opt} used for regular mount points because $conf->{$opt} is not
> necessarily a valid volume ID in this case. This distinction is
> important because is_volume_in_use does in
Am 20.05.25 um 11:08 schrieb Michael Köppl:
> An explicit check for the existence of the storage is added to print a
> warning and continue with the removal of the container without deleting
> the mount point in case the storage does not exist anymore. For other
> errors, the function should still
Signed-off-by: Fiona Ebner
---
I'll apply together with this follow-up if somebody gives me a T-b/R-b
PVE/QemuMigrate.pm | 26 +++---
1 file changed, 11 insertions(+), 15 deletions(-)
diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm
index f72a9d33..5a54f173 1
store test where the backup configuration uses the
short suffixes to still test for that.
Signed-off-by: Fiona Ebner
---
test/cfg2cmd/bootorder-empty.conf | 6 +-
test/cfg2cmd/bootorder-legacy.conf| 6 +-
test/cfg2cmd/bootorder.conf | 6 +-
.../cputy
Applied, thanks!
[1/1] datacenter config: document that the migration network affects
replications
commit: a042611c9d06ea24d02423387005fa97319f9f2e
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailma
The commit title seems to be cut off/incomplete and is missing the "v4".
Sending patches to existing threads is usually done only in exceptional
cases and almost always just for follow-ups. Please send a stand-alone
v5 next time.
Am 14.05.25 um 18:49 schrieb Alexander Abraham:
> I moved the secti
Am 14.05.25 um 16:33 schrieb Thomas Lamprecht:
> Am 13.05.25 um 12:56 schrieb Fiona Ebner:
>> When the service is restarted, the directory will be recreated. The
>> issue is that the /run/pve directory is not just used as the runtime
>> directory of this service, but also
Am 14.05.25 um 20:08 schrieb Thomas Lamprecht:
> Am 13.05.25 um 12:56 schrieb Fiona Ebner:
>> The pve-lxc-syscalld systemd service currently uses /run/pve as a
>> runtime directory. This means, that when the service is restarted, the
>> directory will be recreated. But the
' field refers to the raw usage without factoring out
replication (default: 3). The 'max_avail' value is after factoring out
replication, so using 'bytes_used' in the same calculation would lead
to very confusing results.
Signed-off-by: Fiona Ebner
---
No changes in v2.
src
lly stored user data, i.e. before compression. In the context
of presenting storage usage, this is not the best choice and e.g. in
the ZFS plugin 'used' is preferred over 'logicalused'. Switch to using
'bytes_used' as the basis for the calculation to better match
e
Am 14.05.25 um 11:06 schrieb Fabian Grünbichler:
>> Fiona Ebner hat am 14.05.2025 10:22 CEST geschrieben:
>>
>>
>> Am 13.05.25 um 15:31 schrieb Fiona Ebner:
>>> Signed-off-by: Fiona Ebner
>>> ---
>>> src/PVE/Storage/RBDPlugin.pm | 6 +
Am 13.05.25 um 15:31 schrieb Fiona Ebner:
> Signed-off-by: Fiona Ebner
> ---
> src/PVE/Storage/RBDPlugin.pm | 6 ++
> 1 file changed, 6 insertions(+)
>
> diff --git a/src/PVE/Storage/RBDPlugin.pm b/src/PVE/Storage/RBDPlugin.pm
> index 154fa00..b56f8e4 100644
&g
Signed-off-by: Fiona Ebner
---
src/PVE/Storage/RBDPlugin.pm | 6 ++
1 file changed, 6 insertions(+)
diff --git a/src/PVE/Storage/RBDPlugin.pm b/src/PVE/Storage/RBDPlugin.pm
index 154fa00..b56f8e4 100644
--- a/src/PVE/Storage/RBDPlugin.pm
+++ b/src/PVE/Storage/RBDPlugin.pm
@@ -703,6 +703,12
' field refers to the raw usage without factoring out
replication (default: 3). The 'max_avail' value is after factoring out
replication, so using 'bytes_used' in the same calculation would lead
to very confusing results.
Signed-off-by: Fiona Ebner
---
src/PVE/Storage/RBDPlug
aligns the permissions/owner with the ones /run/pve-cluster has, i.e.
0750 root:www-data.
Signed-off-by: Fiona Ebner
---
We could also opt for 0750 root:root, not sure.
configs/Makefile | 1 +
configs/pve-tmpfiles.conf | 2 ++
2 files changed, 3 insertions(+)
create mode 100644 configs
se the new socket path as soon as it is available. Only use
the old socket path if the old version of pve-lxc-syscalld is still
used. The heuristic to check this is:
1. the new socket path doesn't exist
2. the old socket path exists
3. the old socket path is not a symbolic link
Signed-off-
lost when the service is restarted.
Signed-off-by: Fiona Ebner
---
Versioned breaks for pve-container and pve-manager needed.
etc/pve-lxc-syscalld.service.in | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/etc/pve-lxc-syscalld.service.in b/etc/pve-lxc-syscalld.service.in
(and thus during package upgrade),
since pve-lxc-syscalld doesn't currently implement graceful reloading.
Signed-off-by: Fiona Ebner
---
debian/postinst | 19 +++
1 file changed, 19 insertions(+)
create mode 100644 debian/postinst
diff --git a/debian/postinst b/debian/pos
h requires specifying the new socket path
new pve-lxc-syscalld breaks old pve-manager: /run/pve is not
automatically created in the context of pull metrics yet
I've never used systemd tmpfiles.d before, so that requires extra
scrutiny 0:)
manager:
Fiona Ebner (1):
add tpmfiles.d config to
ABM | CPUID_EXT3_LAHF_LM,
> .features[FEAT_7_0_EBX] =
> CPUID_7_0_EBX_FSGSBASE | CPUID_7_0_EBX_BMI1 |
> CPUID_7_0_EBX_HLE | CPUID_7_0_EBX_AVX2 | CPUID_7_0_EBX_SMEP |
> CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ERMS | CPUID_7_0_EBX_INVPCID |
> CPUID_7_0_EBX_RTM,
Signed-off-by: Fion
Am 12.05.25 um 17:33 schrieb DERUMIER, Alexandre:
> -
>> (for example, they are an "size" option on drive not existing on
>> blockdev for example, I have wrote details on the patch 3/14.)
>
>>> I do think we still need the size (would need to be tested), because
>>> OVMF would get confused when th
Am 12.05.25 um 15:39 schrieb DERUMIER, Alexandre:
> Am 12.05.25 um 12:57 schrieb DERUMIER, Alexandre:
>> for blockdev, do we still use a ceph config file in /var/run for
>> potential others rbd client options ?
>
>>> Not currently, but we can add that later if we consider it worth it.
>>> We
>>> w
Am 09.05.25 um 16:15 schrieb Fiona Ebner:
> Signed-off-by: Fiona Ebner
> ---
>
> for me, it fails with
> VM 102 qmp command 'blockdev-add' failed - LUN is write protected
> but adding as a drive fails. Will need to investigate.
After fixing up my ACLs on the iSCS
The 'disk-size' format in the JSON schema in pve-common was changed to
print more verbose, non-ambiguous disk size suffixes. Adapt the tests
to match how configurations are actually written now.
Bump the corresponding build-dependency in d/control.
Signed-off-by: Fiona Ebner
---
debi
Am 12.05.25 um 14:41 schrieb Fiona Ebner:
> Maybe best is to wait for PVE 9 with this and do a parse+write for all
> guest configs (including their snapshots) in the pve8to9 script? The
> change also breaks backwards migration to a node that doesn't
> understand the new suffix.
The 'disk-size' format in the JSON schema in pve-common was changed to
print more verbose, non-ambiguous disk size suffixes. This requires
adapting some tests here.
Bump the corresponding build-dependency in d/control.
Signed-off-by: Fiona Ebner
---
debian/control
The single-letter suffixes are ambiguous and especially in the context
of disks, the powers of ten are usually used. Use exact suffixes
instead when logging the archive file size.
Signed-off-by: Fiona Ebner
---
PVE/VZDump.pm | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff
lso breaks backwards migration to a node that doesn't
understand the new suffix.
common:
Fiona Ebner (1):
fix #3900: schema: support and prefer sizes with verbose suffixes
{K,M,G,T}iB
src/PVE/JSONSchema.pm | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
manager:
s sense to wait for PVE 9 with this and do a
parse+write for all guest configs (including their snapshots) in
the pve8to9 script?
Signed-off-by: Fiona Ebner
---
src/PVE/JSONSchema.pm | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/src/PVE/JSONSchema.pm
Am 12.05.25 um 12:57 schrieb DERUMIER, Alexandre:
> for blockdev, do we still use a ceph config file in /var/run for
> potential others rbd client options ?
Not currently, but we can add that later if we consider it worth it. We
would need to merge with the storage's already existing ceph.conf and
And here is a small script I used for testing to see if QMP blockdev-add
can deal with the results:
> [I] root@pve8a1 ~# cat test-storage-blockdev.pl
> #!/usr/bin/perl
>
> use strict;
> use warnings;
>
> use JSON;
> use PVE::QemuServer::Monitor qw(mon_cmd);
> use PVE::Storage;
>
> my $vmid = s
Co-developed-by: Alexandre Derumier
Signed-off-by: Fiona Ebner
---
The mon host parsing is adapted from other places where we do that,
but I did not test it with IPv6 yet.
src/PVE/Storage/RBDPlugin.pm | 43
1 file changed, 43 insertions(+)
diff --git a
via NBD or FUSE or external backup
provider restore providing an NBD export for the provider to write to.
This is an early sneak peek to get feedback on the idea itself and to
keep Alexandre in the loop :)
See the individual patches for more comments.
qemu:
Fiona Ebner (1):
block/rbd: ad
Signed-off-by: Fiona Ebner
---
for me, it fails with
VM 102 qmp command 'blockdev-add' failed - LUN is write protected
but adding as a drive fails. Will need to investigate.
src/PVE/Storage/ISCSIDirectPlugin.pm | 17 +
1 file changed, 17 insertions(+)
diff --git
Signed-off-by: Fiona Ebner
---
I'll also submit this upstream.
block/rbd.c | 8
qapi/block-core.json | 4
2 files changed, 12 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
index 24e820d056..53a9a785f0 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -272,6 +2
There intentionally is only handling for absolute paths in the default
plugin implementation. Any plugin requiring more needs to implement
the method itself.
Signed-off-by: Fiona Ebner
---
I discussed this with Fabian off-list. With PVE 9 being a major
release and most popular plugins not using
Signed-off-by: Fiona Ebner
---
src/PVE/Storage/ISCSIDirectPlugin.pm | 14 +++---
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/src/PVE/Storage/ISCSIDirectPlugin.pm
b/src/PVE/Storage/ISCSIDirectPlugin.pm
index 829e0c4..d54dcd8 100644
--- a/src/PVE/Storage
No other plugin activates the storage inside the path() method either.
The caller needs to ensure that the storage is activated before using
the result of path().
Signed-off-by: Fiona Ebner
---
src/PVE/Storage/ESXiPlugin.pm | 1 -
1 file changed, 1 deletion(-)
diff --git a/src/PVE/Storage
Am 09.05.25 um 10:20 schrieb DERUMIER, Alexandre:
> Message initial
> De: Fiona Ebner
> À: "DERUMIER, Alexandre" , pve-
> de...@lists.proxmox.com
> Objet: Re: [pve-devel] [PATCH qemu-server 02/14] blockdev: cmdline:
> convert drive to blockdev syntax
Am 06.05.25 um 16:20 schrieb DERUMIER, Alexandre:
>
> Am 22.04.25 um 13:51 schrieb Alexandre Derumier via pve-devel:
>>
>> fixme:
>> - rbd blockdev don't allow extra options (keyring file for example),
>> do we need to patch qemu ? or write client option in a file ?
>
> I think following upst
Am 06.05.25 um 17:40 schrieb DERUMIER, Alexandre:
>>> Do we need to handle old client (for example, a pve8 doing a live
>>> migrate + lcoal storage migration sending a nbd:// uri to a pve9
>>> with
>>> blockdev support ) ?
>
> Thinking about this, maybe this is a good reason for version guard too
Am 07.05.25 um 10:41 schrieb Fabian Grünbichler:
>> Fiona Ebner hat am 06.05.2025 14:57 CEST geschrieben:
>>> +
>>> + ($storeid) = PVE::Storage::parse_volume_id($volid, 1);
>>> + my $vtype = $storeid ? (PVE::Storage::parse_volname($storecfg,
>>>
1 - 100 of 1935 matches
Mail list logo
