does not really make sense, and if the plugin wants that, it can still
be done, like we do here for the ESXiPlugin
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
[ TL: fix wrong comparison with >= and avoid undef warning if file
does not yet exist at all ]
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
This allows the frontends to translate them and avoids somewhat
duplicated info by having some warnings explicitly (ignored-volumes)
while others are in the warnings array.
By passing along the key and the value the frontend can also show the
warnings in-line, e.g. by marking a disk-entry in a grid as having
potential problems.
Ideally we'd have a central list of known types used for the API
return schema enum and to check when calling the $warn closure, but as
we only got three warnings keep this as is and only add a comment.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
and rename the name of the constant for the FUSE tool, as otherwise it
sounds like this was the mountpoint to be used
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
we should actually keep the scsihw even with eg. linux VMs as they
might not have virtio/... in their initrd and then fail to boot
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
The get_guest_import fn is currently a private/internal API and should
not yet be exposed as a public part of the storage API.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Add a new 'import' content type which will be the corner stone for a
better API and UI integrated way to import virtual guests into Proxmox
VE.
For starters this will be used to implement a ESXi adapter, so that
those VMs can get imported nicely.
Later we want to integrate the OVF/OVA import skeletons we got in
qemu-server to something more usable here.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
[ TL: add more commit message with some background ]
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
The default description doesn't make it clear what the property is
used for in the context of storages.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
If we want to forward to the create_base of the directory plugin while
making that use our $class for the operations that call might do, we
cannot use the -> notation (which would resolve the next actual
implementation) but rather pass the class directly.
But, DirPlugin reuses the create_base method from the base Plugin
method, so we also need to call that, because on direct call notation
the inheritance fallback to super methods isn't available.
Reported in the forum:
https://forum.proxmox.com/threads/95684/post-606535
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
this avoids the need for restarting all services that have pve-storage
modules loaded after the admin installed open-iscsi.
while at it make it a bit more clear that this might die by using
assert in the method name.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
With this patch Proxmox now tries to login to all discovered portals
in case some of them are not logged yet.
In case of multipath configuration when initially configured portal is
missing for some reason Proxmox don't lose iSCSI storage now and can
successfully restore iSCSI connection between reboots.
Signed-off-by: Yuri Konotopov <ykonotopov@gnome.org>
Reviewed-By: Dominik Csapak <d.csapak@proxmox.com>
Tested-By: Dominik Csapak <d.csapak@proxmox.com>
dd supports a 'status' flag, which enables it to show the copied bytes,
duration, and the transfer rate, which then get printed to stderr.
Signed-off-by: Leo Nunner <l.nunner@proxmox.com>
instead of claiming that the storage is not online.
Would've made the issue fixed by b27da68 ("cifs: fix check connection
call") more obvious, because (the UI passes along an empty string for
domain if not set and) the smbclient call returns that status with
> -W ''
in Bookworm.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
adds information for how to decompress isos.
generates the compressor regex from a list of comression formats (to
avoid redundancy)
extends the download_url wtih the functionality to handley compression
for images
Signed-off-by: Philipp Hufnagl <p.hufnagl@proxmox.com>
by not having -W eat the next argument when $scfg->{domain} is defined
but empty.
As reported in the community forum [0], this could lead to issues
adding a storage in the UI.
[0]: https://forum.proxmox.com/threads/44983/post-566842
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Move the warning over to creating and updating storage configs, which
is much less noisy as the constantly called activate storage (e.g.,
pvestatd).
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Since commit 8e623a2930, the inequality
check for content-dirs prints a warning if a content directory path
could not be resolved, i.e., if `abs_path` returns undef. Among other
things, `abs_path` returns undef if the path has an inner (= any but
last) component that does not exist. This can happen for a storage
with content type `iso,vztmpl` and `create-subdirs` set to 0, in case
`template/` does not exist. In this case, the warnings printed by
pvestatd are quite noisy.
As missing content directories are not a problem per se, remove the
warning and just ignore the directory during the inequality check.
Signed-off-by: Friedrich Weber <f.weber@proxmox.com>
This is slightly confusing due to both options, the legacy convoluted
one and the new targeted one, exist, but before the rework we skip if
either of those sub-expressions was true, so doing it needs both to
be true.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
checking the content dirs for clashes via abs_path must be done after
the logic for creating them ran, as abs_path is working on actual
filesystem level, so it will return undf if the directory does not
exist, in which case we then set a hash entry for "undef", and the
next for loop round then resolved again to "undef", resulting in a
false-positive of the check.
Avoid the dangerous "return if" stanzas and reverse them to an actual
if block, which is much safer to adapt. Then move the check for
duplicate content-dir usage after that.
best viewed with white space change ignored: git show -w
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
This makes it possible to add all mount options offered by mount.cifs.
NFS & CIFS now share the options parameter since they use it for the
same purpose.
Signed-off-by: Stefan Hrdlicka <s.hrdlicka@proxmox.com>
[FE: rebase + style fixes]
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Tested-by: Friedrich Weber <f.weber@proxmox.com>
[T: fix merge conflict ]
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
which causes an error "the argument for option '--size' is invalid".
Just round up to the nearest integer to have at least the requested
size. This is similar to what is done for ZFS with d3e3e5d ("When
resizing a ZFS volume, align size to 1M") and makes commands like 'qm
resize 102 scsi1 +0.01G' work.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
It was introduced by commit 4b7dd9d ("allow --allow-shrink on RBD
resize"), but doesn't give a rationale. A mail gives more[0],
indicating that the user also uses the function to shrink images.
However, the volume_resize function is only reachable via the resize
API endpoints for VMs and containers, which have an explicit check to
disallow shrinkage. If somebody really wants to shrink the image, just
let them use the storage's tools directly. Calling into Proxmox VE's
perl functions directly is not supported.
[0]: https://lists.proxmox.com/pipermail/pve-devel/2016-November/024077.html
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
This prevents strange interactions in case the same content directory
is used for multiple content types.
Signed-off-by: Friedrich Weber <f.weber@proxmox.com>
In the previous code, if `create-base-path` was explicitly
set to false, it would be treated the same as if it was
undef, falling through to whatever 'mkdir' was.
Instead, the new options should always be preferred, and the
logic can be simplified to a single line.
Here's the table showing the difference, 'u' being 'undef':
config: mkdir: u 0 1 u 0 1 u 0 1
create: u u u 0 0 0 1 1 1
=========================
mkpath: old: 1 0 1 0 0 1 1 1 1
new: 1 0 1 0 0 0 1 1 1
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
The `mkdir` option has two meanings[0][1] which are split up in `create-path`
and `create-sub-dirs`.
The `create-base-path` option decides if the path to the storage is
automatically created or not.
The `create-subdirs` options decides if the default directory
structure (dump, images, ...) at the storage location is created.
The `mkdir` option is still working but will trigger a warning in the
logs.
As a side effect, this also fixes#3214 because the `create-base-path` option
is now run after the `is_mountpoint` check in the `activate_storage`
method in DirPlugin.pm.
The 'mkpath' command has been moved into a new helper function that
first determines if the conditions to create the path is true, called
'config_aware_base_mkdir'.
[0] https://lists.proxmox.com/pipermail/pve-devel/2020-December/046575.html
[1] https://lists.proxmox.com/pipermail/pve-devel/2020-December/046576.html
Signed-off-by: Aaron Lauterer <a.lauterer@proxmox.com>
Reviewed-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
