IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
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>
The `tmpfilename` is generated by pve-http-server and always adheres
to this pattern, so make this explicit to prevent confusion and have
a more complete JSON API schema, useful for e.g., the API viewer.
Signed-off-by: Friedrich Weber <f.weber@proxmox.com>
[ T: slightly extended commit message ]
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
this keeps the mount sources consistent with previous versions
without this patch there is a small regression, which leads to the
storage not being recognized as being mounted on upgrade:
* pvestatd in older version mount the storage with out trailing /
```
//cifsstore/ISO on /mnt/pve/cifsstore type cifs...
```
* the cifs_is_mounted helper does not recognize it as being mounted
(as the source now has a / in the end)
* attempting to mount leads to
```
mount error(16): Device or resource busy
```
noticed after upgrading and having a cifs storage mounted
Signed-off-by: Stoiko Ivanov <s.ivanov@proxmox.com>
`pvesh get /nodes/{node}/storage/{storage}/content/{volume}` failed for
several storage types, because the respective storage plugins returned
only the volumes `size` on `volume_size_info` calls, while also the format
is required.
This patch fixes the issue by returning also `format` and where possible `used`.
The issue was reported in the forum:
https://forum.proxmox.com/threads/pvesh-get-nodes-node-storage-storage-content-volume-returns-error.123747/
Signed-off-by: Christian Ebner <c.ebner@proxmox.com>
[ T: fixup white space error ]
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Allow a dot as long as its not followed by another dot and enforce
max component and (a reduced) max path length checking already at
schema level.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
CIFS/SMB supports directly mounting subdirectories, so it makes sense to
also allow the --subdir parameter for these storages. The subdir
parameter was moved from CephFSPlugin.pm to Plugin.pm, because it isn't
specific to CephFS anymore.
Signed-off-by: Leo Nunner <l.nunner@proxmox.com>
by simply doing a ping with the expected port as a fallback when the
rpcinfo command fails.
The timeout was chosen to be 2 seconds, because that's what the
existing callers of tcp_ping() in the iSCSI and GlusterFS plugins use.
Alternatively, the existing check could be replaced, but that would
1. Dumb down the check.
2. Risk breakage in some corner case that's yet to be discovered.
3. It would still be necessary to use rpcinfo (or dumb the check down
even further) in case port=0; from 'man 5 nfs' about the NFSv4 'port'
option:
> If the specified port value is 0, then the NFS client uses the NFS
> service port number advertised by the server's rpcbind service.
Reported in the community forum:
https://forum.proxmox.com/threads/118466/post-524449https://forum.proxmox.com/threads/120774/
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
The only caller where $running can even be truthy is QemuServer.pm's
qemu_volume_snapshot_delete(). But there, a check if snapshots should
be done with QEMU is already made and the storage function is only
called if snapshots should not be done with QEMU (like for TPM drives
which are not attached directly). So rely on the caller and do not
return early to fix removing snapshots in such cases.
Even if a stray call ends up here (can happen already by changing the
krbd setting while a VM is running to flip the above-mentioned check
and the early return check removed by this patch), it might not even
be problematic. At least a quick test worked fine:
1. take snapshot via a monitor command in QEMU
2. remove snapshot via the storage layer
3. create a new file in the VM
4. take a snapshot with the same name via monitor command in QEMU
5. roll back to the snapshot
6. check that the file in the VM is as expected
Using the storage layer to take the snapshots and QEMU to remove the
snapshot also worked doing the same test. Even if it were problematic,
the check in qemu-server should rather be improved then.
(Trying to issue a snapshot mon command for a krbd-mapped image fails
with an error on the other hand, but that is also not too bad and not
relevant to the storage code. Again, it rather would be necessary to
improve the check in qemu-server).
The fact that the pve-container code didn't even pass $running is the
reason removing snapshots worked for containers on a storage with krbd
disabled (the pve-container code calls map_volume() explicitly, so
containers can work regardless of the krbd setting in the storage
configuration; see commit 841fba6 ("call map_volume before using
volumes.") in pve-container).
For volume_snapshot(), the early return with $running was already
removed (or rather the relevant logic moved to QemuServer.pm) in 2015
by commit f5640e7 ("remove running from Storage and check it in
QemuServer"), even before krbd support was added in RBDPlugin.pm.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Reviewed-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
To be correct in all cases, it's still necessary to filter by "pool"
in zfs_parse_zvol_list(), because $scfg->{pool} could be e.g.
'foo/vm-123-disk-0' which looks like an image name and would pass the
other "should skip"-checks in zfs_parse_zvol_list().
No change in the result of zfs_list_zvol() is intended.
Suggested-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
The 'pool' property in the result of zfs_parse_zvol_list() was not
used for anything else.
No functional change is intended.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Avoids the need for the fallback for the (only existing) caller.
Note that the old my $list = (); is a rather intransparent way of
assigning undef.
Suggested-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Rename the "dirs" parameter to "content-dirs". Switch from a "vtype:/dir"
format to "vtype=/dir", and remove the misleading error message talking
about "absolute" paths. One might expect these to be absolute over the
whole system, while in reality they are relative to the mountpoint of
the storage.
Signed-off-by: Leo Nunner <l.nunner@proxmox.com>
Allowing overrides for the default directory locations seems to
integrate rather well into the existing system. Custom locations
are specified using the "dirs" parameter as a comma-separated list
of "vtype:/location" values.
For now, the option has been enabled for the Directory, CIFS and NFS
backends.
Signed-off-by: Leo Nunner <l.nunner@proxmox.com>
There is no caller using $cache and the same $storeid multiple times,
so there is no need to keep the cache.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
The plugin for remote ZFS storages currently also uses the same
list_images() as the plugin for local ZFS storages. There is only
one cache which does not remember the target host where the
information originated.
This is problematic for rescan in qemu-server:
1. Via list_images() and zfs_list_zvol(), ZFSPlugin.pm's zfs_request()
is executed for a remote ZFS.
2. $cache->{zfs} is set to the result of scanning the images there.
3. Another list_images() for a local ZFS storage happens and uses the
cache with the wrong information.
The only two operations using the cache currently are
1. Disk rescan in qemu-server which is affected by the issue. It is
done as part of (or as a) long-running operations.
2. prepare_local_job for pvesr, but the non-local ZFS plugin doesn't
support replication, so it cannot trigger there. The cache only
helps if there is a replicated guest with volumes on different
ZFS storages, but otherwise it will be less efficient than no
cache or querying every storage by itself.
Fix the issue by making the cache $storeid-specific, which effectively
makes the cache superfluous, because there is no caller using the same
$storeid multiple times. As argued above, this should not really hurt
the callers described above much and actually improves performance for
all other callers.
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
This essentially reverts commit c9bd3d2 ("fix #1123: modify NVME
device path for SMART support").
The man page for smartctl states
> Use the forms "/dev/nvme[0-9]" (broadcast namespace) or
> "/dev/nvme[0-9]n[1-9]" (specific namespace 1-9) for NVMe devices.
so it should be fine to pass the path with the specific namespace to
smartctl.
But that text was already present in the man page of version 6.5,
which is the version the commit c9bd3d2 talks about. It might be that
it was necessary to drop the specific namespace for the version
backported from Stretch to Jessie (the bug report mentions that that
version was used[0]), but it's not quite clear.
With current versions, passing in the path with the specific namespace
did work as expected[1], even on a device with multiple namespaces set
up tested locally. In PBS, the path queried via
udev::Device::from_syspath("/sys/block/{name}") is passed to smartctl
and that also included the specific namespace on the systems I tested
with a short script.
So pass the full path to make things a little bit simpler and to avoid
potential future issues like bug #2020[2].
[0]: https://bugzilla.proxmox.com/show_bug.cgi?id=1123#c3
[1]: https://forum.proxmox.com/threads/113962/post-493185
[2]: https://bugzilla.proxmox.com/show_bug.cgi?id=2020
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
This reverts commit c3442aa554.
Nowadays, relying on 'readlink /sys/block/nvmeXnY/device' won't always
lead to the correct device, as reported in the community forum[0],
where it results in '../../nvme-subsys0' and there's no matching entry
under '/dev/'.
Since Linux kernel 5.4, in particular commit 733e4b69d508 ("nvme:
Assign subsys instance from first ctrl"), the problematic situation
from bug #2020 shouldn't happen anymore.
Stated more clearly by the commit's author here[1]:
> Indeed, that commit will make the naming a bit more sane and will
> definitely prevent mistaken identity. It is still possible to
> observe controllers with instances that don't match their
> namespaces, but it is impossible to get a namespace instance that
> matches a non-owning controller.
The only other user of get_sysdir_info() doesn't use the 'device'
entry, so reverting that part is fine too.
[0] https://forum.proxmox.com/threads/113962/
[1] https://github.com/linux-nvme/nvme-cli/issues/510#issuecomment-552508647
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
Tested-by: Stefan Hrdlicka <s.hrdlicka@proxmox.com>
Previously, calling with e.g. $storage_list = [undef] would lead to an
early return of $override and not consider the limit from
datacenter.cfg.
Refactoring the bandwidth limit handling for migration introduced
calls such as described above, which broke applying the limit from
datacenter.cfg for VM RAM/state migration.
Reported in the community forum:
https://forum.proxmox.com/threads/37920/post-513005
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
file-restore has a 'timeout' parameter and if that is exceeded, returns
an error with the http code 503 Service Unavailable
When the web ui encounters such an error, it retries the listing a few
times before giving up.
To make use of these, add the 'timeout' parameter to the new
'extraParams' of 'file_restore_list'.
25 seconds are chosen because it's under pveproxy 30s limit, with a bit
of overhead to spare for the rest of the api call, like json decoding,
forking, access control checks, etc.
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
since commit
ba690c40 ("file-restore: remove 'json-error' parameter from list_files")
in proxmox-backup, the file-restore binary will return the error as json
when called with '--output-format json' (which we do in PVE::PBSClient)
here, we assume that 'file-restore' will fail in that case, and we try
to use the return value as an array ref which fails, and the user never
sees the real error message.
To fix that, check the ref type of the return value and act accordingly
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
