--update-version (intended for use by script maintainer only) will derive
version number from the output of "git describe" and set the SCRIPT_VERSION
variable inside the script.
--version will display a version number message and exit immediately thereafter.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
When updating the Backport tracker issue in Redmine, upon success the status
code should be 2**.
A status code of 4** indicates a problem with authentication.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
The initial implementation sent events to the
kubernetes cluster Ceph is running under managed
by rook-ceph. This patch extends this support to
include sending events to an external kubernetes
cluster, that may just be consuming ceph resources
Additional docs added to help people use the
module, either with or without rook-ceph.
It also address a bug where the module was loaded
outside of a rook-ceph environment.
Fixes: https://tracker.ceph.com/issues/41737
Signed-off-by: Paul Cuzner <pcuzner@redhat.com>
Increasing osd_object_clean_region_max_num_intervals to track more
clean regions, resulting in more partial recovery.
Signed-off-by: Neha Ojha <nojha@redhat.com>
* refs/pull/29711/head:
doc: Update documentation for LazyIO
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
* refs/pull/30251/head:
doc: Update requirements for using CephFS
Reviewed-by: Rishabh Dave <ridave@redhat.com>
Reviewed-by: Douglas Fuller <dfuller@redhat.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
* refs/pull/30627/head:
qa/suites/rados/verify: whitelist MON_DOWN when using valgrind
qa/suits/rados/basic/tasks/rados_api_tests: pgs can go degraded
Reviewed-by: Neha Ojha <nojha@redhat.com>
common/secret.c: don't pass uninitialized stack data to the kernel
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Sage Weil <sage@redhat.com>
* refs/pull/30626/head:
doc: document recover_session= option in mount.ceph manpage
doc: update documents on using kcephfs
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
* refs/pull/30416/head:
client: add procession of SEEK_HOLE and SEEK_DATA in lseek.
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
ceph_unarmor() returns the number of bytes decoded, which can be (and
usually is) smaller than the size of the payload array.
set_kernel_secret() has behaved this way ever since it was introduced
in commit bee85518e2 ("mount.ceph: Use kernel key management API when
possible."). The reason it didn't cause problems in the kernel is that
the encoding includes the actual length of the secret and there is no
check for the end of the supplied payload (see ceph_key_preparse() in
net/ceph/crypto.c).
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Overhaul the fstab.rst and kernel.rst files with more relevant examples
now that the mount helper does autodiscovery. Add a preamble to
kernel-features.rst and add a warning about the deprecation of inline
data.
Fixes: https://tracker.ceph.com/issues/42025
Signed-off-by: Jeff Layton <jlayton@redhat.com>
When constructing the snapshot path, the components are a mix of
byte and string type objects. The 'os.path.join()' method needs
the components to be of the same type. Hence convert all components
to byte type objects.
Introduced-by: 3d63cd947f
Fixes: https://tracker.ceph.com/issues/42096
Signed-off-by: Ramana Raja <rraja@redhat.com>