if PVC is created from a snapshot, external-provisioner
expects the volume ContentSource to be set in Create
VolumeResponse
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit a4f6a3407f)
currently, we are making use of host path directory
to store the provisioner socket, as this
the socket is not needed by anyone else other than
containers inside the provisioner pod using the
empty directory to store this socket is the best option.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit fbda8cc4ca)
On systems with SELinux enabled, non-privileged containers
can't access data of privileged containers. Since the socket
is exposed by privileged containers, all sidecars must be
privileged too. This is needed only for containers running
in daemonset as we are using bidirectional mounts in daemonset
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit e0cc7740f6)
if the directory is not mounted return nil
during umount of mountPoint
Discard error if error is os.IsNotExist
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit dcafdb519e)
as part of https://github.com/ceph/ceph-csi/pull/715/
we have updeated the release-v1.2.0 branch to
push v1.2.2 images as the release is done reverting back
the image tag to push canary image.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
Linux kernel 4.17.0 adds support for quota with CephFS. Without quota,
it is not possible to fullfill the requirements of the CSI Spec and
guarantee sufficient space on the filesystem for a volume. With this in
mind, usage of the kernel client is only allowed with kernel 4.17.0 or
newer.
However, some Linux vendors backport features and patches to their
Enterprise products. These kernels may have an older version, but do
support quota. One of these is the kernel that comes with RHEL-7.7.
By comparing the current running version of the Linux kernel against
known versions that support quota, we can now automatically decide to
use the kernel client, or not.
Note that this does not change the 'forcekernelclient' parameter. The
parameter is still available and can be used for kernels that are not in
the 'known to support quota list'. Or users can pass the parameter to
use a CephFS kernel client that does not support quota.
Signed-off-by: Niels de Vos <ndevos@redhat.com>
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 290beb4dda)
From master and release-v1.2.0 branch we will be
pushing the canary charts. when we are doing the release
i.e v1.2.2 from release-v1.2.0 branch we will replace
all canary to released tag v1.2.2 and push helm charts
for v.1.2.2 and will revert back the changes in
release-v1.2.0 branch push canary tagged charts(
this will be same logic as pushing the container image)
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 50268d5947)
If the RBD user does not have permissions to talk to the
Ceph MGR, it should gracefully fallback to the slower foreground
image deletion.
Fixes: #677
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit a274b19bfa)
updated cephfs registration directory
name to match with rbd implementaion
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 9287948991)
this time out value to 150s or higher. The higher timeout value can help to reduce the
load of our backend ceph cluster and also can avoid throttling issues at sidecars to an extent.
Fix# #602
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
(cherry picked from commit 1efdf14ac5)
Skip zeroing of the journal on freshly created images. As only dynamic
PVs are supported, it is fine to assume that existingFormat == "" image
has never been mapped and written to before.
lazy_itable_init is enabled by default, passing it for consistency with
lazy_journal_init.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
(cherry picked from commit 9a61fa0c0a)
as per the examples rules in
https://doc.mergify.io/examples.html, string are
surrounded by `""`, it better to follow same
format here also.
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit ed364cf6ea)
Improve the error message if
secrets are not provided in request
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit b8568a5bb9)
rootfs dependency was removed from rbd
by removing support for `nsenter`, This
PR removed the `/` mount from provisioner
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 3d0cba1931)
However further testing shows that, ext4 should be
the default or preferred fs for RBD devices.
This patch bring that change
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
(cherry picked from commit b44a81bdbc)
Currently rbd CSI plugin uses formatAndMount of
mount.SafeFormatAndMount. This does not allow to pass or use
specific formatting arguments with it. This patch introduce
RBD specific formatting options with both xfs and ext4,
for example: -E no-discard with ext4 and -k option with
XFS to boost formatting performance of RBD device.
Signed-off-by: Humble Chirammal <hchiramm@redhat.com>
(cherry picked from commit 0e6617e1ff)
if any on going opearation is seen,we
have to return Abort error message
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 6aac399075)
rook master is not good enough to
depend on for E2E testing, swithing to v1.1.0
for E2E
Signed-off-by: Madhu Rajanna <madhupr007@gmail.com>
(cherry picked from commit 18151e9ca8)