In this case the current stack trace points to the line
where the context was created. Instead the stack should be
captured when the defer is running so the return path to
the defer call is also part of the stack.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
These were added in 0b5a315c22, because the
continuity/fs package did not provide a Windows implementation. They
were upstreamed in [continuity@3cbda8c], which is part of continuity v0.4.4,
so we can remove the implementation here.
[continuity@3cbda8c]: 3cbda8c24b
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
If container exits with error and has invoked OOMKiller
mark the origin error as ENOMEM so that it can be detected
on the client side.
gRPC will set ENOMEM as codes.ResouceExhausted based on #5182
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
Commit 4b85f11164 migrated the use of the
userns package to the github.com/moby/sys/user module.
After further discussion with maintainers, it was decided to move the
userns package to a separate module, as it has no direct relation with
"user" operations (other than having "user" in its name).
This patch migrates our code to use the new module.
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
The userns package in libcontainer was integrated into the moby/sys/user
module at commit [3778ae603c706494fd1e2c2faf83b406e38d687d][1].
The userns package is used in many places, and currently either depends
on runc/libcontainer, or on containerd, both of which have a complex
dependency tree. This patch is part of a series of patches to unify the
implementations, and to migrate toward that implementation to simplify
the dependency tree.
[1]: 3778ae603c
Signed-off-by: Sebastiaan van Stijn <github@gone.nl>
If context is canceled before the process is ready
then kill goroutine returns early because there is nothing
to kill. But the process may still start after this and
that case remain running without cancellation. Fix is to skip
cancellation only if the run goroutine is ended, as then the
process will not be started.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
Historic layer limit for Docker images is 127. Because in
overlayfs mounting 127 layers usually reaches the page size
limit of mount options in Linux kernel, there is special code
to work around the limitation.
This custom code was used for rootfs of container because
runc takes rootfs as a directory path, meaning buildkit needs
to mount it and then pass the path. For non-rootfs mounts
runc takes them as direct mount configuration and performs
the mount itself. As runc does not have this special way to
mount long overlayfs mounts it will perform the mount with clipped
options what will fail in some way in kernel depending on the
precise cutoff point.
Workaround is to detect when the mount passed to runc is too
long for runc to mount it itself and it that case let
BuildKit mount it and in runc perform bind of the BuildKit mount.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
The source path changed in v0.13 and there are reports
that new path can cause error on starting a container. While
this is investigated, check for missing path and make
mounting optional like it was in v0.12.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
While submount paths were already validated there are some
cases where the parent mount may not be immutable while the
submount is created.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
(cherry picked from commit 2529ec4121bcd8c35bcd96218083da175c2e5b77)
Running interactive container APIs was done by giving
the gateway implementation access to worker controller
directly, but it should be passed with a build job instead.
Signed-off-by: Tonis Tiigi <tonistiigi@gmail.com>
(cherry picked from commit 0971dffaab93d91e51af984b44c745b35b3c5b4d)
Like CRI, we should support custom shim paths. We can just add this as a
new config field "path" under the runtime object.
While "name" can be either the name of a custom shim (from which a path
is derived), or a path itself, if configured as a path, no options can
be provided - this is because to derive the type for the options struct,
we need the name to be a member of a well-known set (currently runc and
hcs).
With this patch, it's now possible to configure a runtime with a custom
shim at a non-default path, and include it's options (which was
previously not possible to do in buildkit).
Signed-off-by: Justin Chadwell <me@jedevc.com>
