[release/1.7 backport] Mount snapshots on Windows #8616
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The WCOW layer support does not support creating sandboxes with no parent. Instead, parentless scratch layers must be layed out as a directory containing only a directory named 'Files', and all data stored inside 'Files'. At commit-time, this will be converted in-place into a read-only layer suitable for use as a parent layer. The WCOW layer support also does not deal with making read-only layers, i.e. layers that are prepared to be parent layers, visible in a read-only manner. A bind-mount or junction point cannot be made read-only, so a view must instead be a small sandbox layer that we can mount via WCOW, and discard later, to protect the layer against accidental or deliberate modification. Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Using symlinks for bind mounts means we are not protecting an RO-mounted layer against modification. Windows doesn't currently appear to offer a better approach though, as we cannot create arbitrary empty WCOW scratch layers at this time. For windows-layer mounts, Unmount does not have access to the mounts used to create it. So we store the relevant data in an Alternate Data Stream on the mountpoint in order to be able to Unmount later. Based on approach in containerd#2366, with sign-offs recorded as 'Based-on-work-by' trailers below. This also partially-reverts some changes made in containerd#6034 as they are not needed with this mounting implmentation, which no longer needs to be handled specially by the caller compared to non-Windows mounts. Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com> Based-on-work-by: Michael Crosby <crosbymichael@gmail.com> Based-on-work-by: Darren Stahl <darst@microsoft.com>
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
This is necessary on Windows, as it's not possible to delete a snapshot while it is still mounted, even if the mount-point has been deleted. Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Filesystem permissions and ownership are not modifiable via an image mount. Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
NTFS, when presented with an all-caps filename, assumes you are just being loud for no reason, and instead stores an all-lower-case filename. Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
Signed-off-by: Paul "TBBle" Hampson <Paul.Hampson@Pobox.com>
The bind filter supports bind-like mounts and volume mounts. It also allows us to have read-only mounts. Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
fstest.CheckDirectoryEqual checks if any files in the diff matches a list of known metadataFiles. This only happens if we specify the initial layer as the first parameter and the mutated layer as the second. This also enables the read-only view checks, as the bind filter allows us to mount a layer as ro. Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Update dependencies and remove the local bindfilter files. Those have been moved to go-winio. Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
As opposed to a writable layer derived from a base layer, the volume path of a base layer, once activated and prepared will not be a WCIFS volume, but the actual path on disk to the snapshot. We cannot directly mount this folder, as that would mean a client may gain access and potentially damage important metadata files that would render the layer unusabble. For base layers we need to mount the Files folder which must exist in any valid base windows-layer. Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
* Improve error messages
* remove a check for the existance of unmount target. We probably
should not mask that the target was missing.
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
|
Hi @gabriel-samfira. Thanks for your PR. I'm waiting for a containerd member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
|
The diff looks reasonable for backport. Keep the dependency version pinned to a tag though. |
gabriel-samfira
force-pushed
the
backport-8043
branch
from
6e46961
to
e894b50
Compare
dcantah
approved these changes
Jun 1, 2023
|
I updated the continuity dependency to a tagged version. The failure seems unrelated. |
Signed-off-by: Gabriel Adrian Samfira <gsamfira@cloudbasesolutions.com>
gabriel-samfira
force-pushed
the
backport-8043
branch
from
e894b50
to
313c226
Compare
dmcgowan
approved these changes
Jun 1, 2023
kzys
approved these changes
Jun 2, 2023
|
Seems to have caused a regression: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There is an effort to enable native Windows support in buildkit and the functionality that merged in #8043 is a large dependency for that work.
The buildkit project maintainers understandably prefer to stick to stable releases of containerd.
I realize that backports are not normally done for features, but considering there is no defined timeframe for the next major version release of containerd, it would help tremendously if we could backport this branch and unblock a bunch of PRs in buildkit.