And rename it to ChownClearSetuid, which will test the chown only.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Write/truncate by privileged user leaves the suid and sgid and by
unprivileged user will clear them.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Both the STATX_MODE and SETATTR_MODE equal 1, so the test didn't
fail.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Currently the ftruncate will always clear the suid/sgid even the
privileged users are truncating the file. This isn't consistent with
all the other existing filesystems in Linux.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
Currently the write will clear the suid/sgid always even the
privileged users are writing the file. This isn't consistent with
all the other existing filesystems in Linux.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
There is no Posix item requires that we should clear the suid/sgid
in fallocate code path but this is the default behaviour for most of
the filesystems and the VFS layer. And also the same for the write
code path, which have already support it.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
The fallocate path will try to clear the suid/sgid if a unprivileged
user changed the file. And the VFS will set the ATTR_MODE and at the
same time by clearing the suid/sgid mode bits.
There is no Posix item requires that we should clear the suid/sgid
in fallocate code path but this is the default behaviour for most of
the filesystems and the VFS layer. And also the same for the write
code path, which have already support it.
Fixes: https://tracker.ceph.com/issues/58680
Signed-off-by: Xiubo Li <xiubli@redhat.com>
rgw: add SiteConfig to load/create zonegroup configuration
Reviewed-by: Daniel Gryniewicz <dang@redhat.com>
Reviewed-by: Matt Benjamin <mbenjamin@redhat.com>
if a specific rgw_service_provider_name is not configured, add a default
Server response header that includes the ceph release name:
Server: Ceph Object Gateway (reef)
Signed-off-by: Casey Bodley <cbodley@redhat.com>
ajsf json schema library for angular doesn't seem to be actively
maintained. Instead, fromly is a well maintained replacement with extra
stuff like validators builtin, support for json schemas, custom
components, etc...
Textareas weren't supported on ajsf, therefore, it made sense to move to
this dep instead.
Signed-off-by: Pere Diaz Bou <pdiazbou@redhat.com>
Signed-off-by: Nizamudeen A <nia@redhat.com>
use delete_marker from the DeleteOp result, not from the sal::Object
Fixes: https://tracker.ceph.com/issues/59048
Signed-off-by: Casey Bodley <cbodley@redhat.com>
* refs/pull/50493/head:
qa: fixup quota values exceeding 64 bit signed int value
qa, cleanup: no need to specify human readable quota values in comment
Reviewed-by: Dhairya Parmar <dparmar@redhat.com>
* refs/pull/48382/head:
test: add test to validate attrs for snapdir after snap operations
client: update {m,c}time and change attr for snapdir from last modification time of a snap realm
mds: update last modification time and change attr for a snap realm node
mds: include last modification time and change attr of a snap realm node in snap trace reply
mds: use change_attr from closest ancestor when initializing a new snaprealm
mds: initialize last_modified to current timestamp for root inode snaprealm
mds: track last modification time and change attr for a snap realm node (srnode)
Reviewed-by: Xiubo Li <xiubli@redhat.com>
Reviewed-by: Milind Changire <mchangir@redhat.com>
* refs/pull/50049/head:
qa: add cephfs_mirror test case to check root dir modes
cephfs_mirror: sync snap dir root mode with remote dir
Reviewed-by: Dhairya Parmar <dparmar@redhat.com>
Reviewed-by: Venky Shankar <vshankar@redhat.com>
Attempting to enable mirroring on an an image within a namespace bails
out without any useful msg.
$ rbd mirror image enable mi_pool/ns_0/big snapshot
2021-11-17T08:33:21.321+0000 7f4b36f9c2c0 -1 librbd::api::Mirror:
image_enable: cannot enable mirroring in the current pool mirroring mode
Fixes: https://tracker.ceph.com/issues/58895
Signed-off-by: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
Disambiguate and improve the English language in
doc/rados/operations/bluestore-migration.rst up to but not including the
section called "Whole Host Replacement".
Co-authored-by: Cole Mitchell <cole.mitchell.ceph@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
The test does the following:
setfattr -n ceph.quota.max_bytes -v 8388608Ti .
This failes since 8388608Ti exceeds 64 bit signed integer value.
Change that to the approprivate value and expect faliure when
quota is set for this exceeded value.
Fixes: http://tracker.ceph.com/issues/55940
Introduced-by: 7330651482
Signed-off-by: Venky Shankar <vshankar@redhat.com>
Disambiguate and improve the English language in
doc/rados/operations/bluestore-migration.rst up to but not including the
section called "Whole Host Replacement".
Co-authored-by: Cole Mitchell <cole.mitchell.ceph@gmail.com>
Signed-off-by: Zac Dover <zac.dover@proton.me>
