gofumpt is a superset of gofmt, enabling some more code formatting
rules.
This commit is brought to you by
gofumpt -w .
Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
unprivileged users cannot use the trusted.* xattrs. Since for
rootless we always mount overlay with userxattr, we can just check if
running in rootless mode and use user.* instead of trusted.*.
Closes: https://github.com/containers/podman/issues/9936
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
This way the error messages are more informative.
Convert all the users accordingly (assuming there are no users outside
of this repo).
Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
This should avoid the following panic in Lgetxattr:
panic: runtime error: makeslice: len out of range
Fixes: 0da4bc60b3 ("pkg/archive: strictly handle errors reading xattrs")
See: https://travis-ci.org/containers/buildah/jobs/617652963
Signed-off-by: Zac Medico <zmedico@gmail.com>
Preservation of user.* xattrs is supported by most (if not all) other
drivers, and it's especially useful for containers that run under a
PaX kernel where "user.pax.flags" is used to store PaX flags.
Signed-off-by: Zac Medico <zmedico@gmail.com>
Some platforms store s.Rdev as unit32, translate to unit64 so it
will compile.
Also add function to make it clear that the object is a whiteout object.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Expose reading and writing ID mapping in the archive and chrootarchive
packages, and in the driver interface. Generally this means that
when computing or applying diffs, we need to have ID mappings passed in
that are specific to the layers we're using.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
This fixes compilation on macOS, at least, and by code inspection could
work on Windows as well.
Fixes#160.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
If the overlay filesystem implements an optimization that landed in
97c684cc911060ba7f97c0925eaf842f159a39e8, and in the mainline kernel in
4.10, directories created in merged directories are marked as opaque by
the kernel to let the kernel know that it needn't bother looking at
other layers when reading the contents of that directory.
This means that, when generating a diff for an upper directory, we can't
treat the presence of an opaque attribute as enough of an indication
that a layer diff needs to include whiteout for a directory of the same
name from a lower layer.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
When deciding whether to mark a file as added or modified, we check for
the presence of an item with the same path in a layer from which the
current later might have inherited a different version of the file.
This can be tripped up by overlay-style whiteout entries, which have the
same name as the file which they mark as having been removed. Also make
sure that we don't mistakenly count, as inherited, a file with the same
name that was in a lower layer, but which was removed by an intermediate
layer.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
Had to vendor in a new version of golang.org/x/net to build
Also had to make some changes to drivers to handle
archive.Reader -> io.Reader
archive.Archive -> io.ReadCloser
Also update .gitingore to ignore emacs files, containers-storage.*
and generated man pages.
Also no longer test travis against golang 1.7, cri-o, moby have also
done this.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Rename the library module and CLI wrapper.
Rename daemon/graphdriver to drivers.
Catch up vendoring to match modules we've pruned.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
If we tear through a few layers of abstraction, we can get at the inodes
contained in a directory without having to stat all the files. This
allows us to eliminate identical files much earlier in the changelist
generation process.
Signed-off-by: Burke Libbey <burke@libbey.me>
Kir Kolyshkin