Some file systems don't support renameat2's RENAME_NOREPLACE
flag (eg. ZFS), at the some time, some other file systems
don't support hardlinks via link (eg. vfat, cifs), so we now
try both: first the rename (since it's more efficient), then
link+unlink for the rest.
If both fail, the file system is simply not supported for
our purposes anyway...
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
we are not really compatible with pbs1, needs a stable-1 branch if we
need to backport something
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
this major release still needs to have an incompatible, the next one can
drop setting a protocol client-side, and the one after that can remove
the protocol handling on the server side.
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
we do not support websocket subprotocols, but for compatibility with
current clients (novnc, pve-xtermjs) we have to reply with the one requested,
else this is a protocol error and browsers will error out
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
we never actually sent text frames, nor did any client request them.
Also, no validity check ever ocurred, so technically it was against
the spec.
Simply remove the code handling sending out text frames. If we need
to actually handle that, we can always create a 'WebSocketStringWriter'
or similar.
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
it's a 'http' utility crate
* the extra 'http' module is redundant
* the module name 'helpers' as a public path doesn't say
much about what it does
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
this is just a (rather HTTP specific) wrapper, so put it into a
'wrapper' module for now.
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
with test1/3 we only tested the regex, which is now gone
transform the tests to valid emails, so we test this also
and rename them to know what they should test
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
by removing the regex check here, that is responsibility of the caller
this is ok since we pass the args directly and not via shell, so
command injection should not be possible
Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
The :vis macro matcher has been stable for a long time now.
Let's use it, and replace the recursion with a plus pattern.
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
