There are two fields I'm worried about: shared namespaces and pod
containers. Both are generated via loops over maps and are thus
non-deterministic in ordering. Throw a sort on each to fix the
order so we can actually diff `podman pod inspect` output.
Signed-off-by: Matt Heon <mheon@redhat.com>
The `*os.File` returned by `pathrs.OpenatInRoot` needs to
be closed before returning from `openSymlinkPath`
Signed-off-by: Christophe Fergeau <cfergeau@redhat.com>
In function rm variable `report` might be initialized as nil as a result
of call
`registry.ImageEngine().Remove(registry.Context(), args, imageOpts)`.
Then, there is a call `registry.SetExitCode(report.ExitCode)` without
explicit nil check before. Check `len(rmErrors) > 0` doesn't guarantee
that report is a non-nil value.
So such call may lead to nil deref.
This commit adds check `report` for nil before its dereference.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Signed-off-by: Mikhail Dmitrichenko <m.dmitrichenko222@gmail.com>
There is a couple of newGenericDecompressor function usages, where
returned possibly non-nil `err` is not checked before dereferencing
returned decompressor. It may lead to nil ptr dereferencing.
This commit adds check for `err` to prevent dereferencing potentially
nullable decompressor.
Found by Linux Verification Center (linuxtesting.org) with SVACE
Signed-off-by: Mikhail Dmitrichenko <m.dmitrichenko222@gmail.com>
This was not working when emulating Linux container images on FreeBSD.
The code to handle host devices on FreeBSD relies on the container
having a devfs mount. Unfortunately, the Linux emulation code which adds
this was happening after the host device handling. This changes the
logic so that host device management happens after Linux emulation.
Signed-off-by: Doug Rabson <dfr@rabson.org>
The WSL machine start was using the function FindExecutablePeer that
ignores user configuration (helper_binaries_dir). FindHelperBinary
instead is used when starting the machine for the rest of the providers
and honors user configuration.
This commit requires 4877783c37
Signed-off-by: Mario Loriedo <mario.loriedo@gmail.com>
The code checks for isPathOnVolume and isPathOnMount so we can just use
the SecureJoin here directly to check for path existance.
Then instead of walking symlinks and trying to guess if they are on a
mount just assume if it is a link (path is different from the normal
joined one) then don't error out early and let the OCI runtime deal with
it. The runtime does produce a less readable error but it still fails
and we have much less fragile code.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
MkdirAll can fail with EEXIST when the path is a symlink and the target
doesn't exist. As such we should ignore the error.
Note there is something fundemantal wrong here with the path access as
it is following the symlink to the host, however it is only for a
stat() so it is not an security issue here.
Fixes: 637c264e2e ("fix issues found by nilness")
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
This bumps Buildah to v1.42.2 and runc to v1.3.4.
This fixes the regressions in run that were created during the
recent fixes for the thre CVEs.
Signed-off-by: tomsweeneyredhat <tsweeney@redhat.com>
There doesn't seem any reason why the system commands should not join
the userns. In particular the main commands use ParentNSRequired and
UnshareNSRequired when they don't want to be joined to the main userns.
Since the system command don't set these the go code does the join and
re-exec anyway so might as well use the shortcut to speed that up.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
When trying to join the conmon pid to recreate the pause process based
on the namespace it can be that the pid is no longer valid, i.e. when
conmon crashed or was killed.
Currently we have a big issue that can be reproduced using:
$ podman run -d quay.io/libpod/testimage:20241011 sleep 100
$ killall -9 conmon
$ killall catatonit
All commands would fail as we keep trying to rejoin the namespace of the
non existing conmon process.
So to address that fall back to creating a new namespace if we fail to
join the conmon pids.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Just a minor improvement as we know the size needed for the slice we can
allocate it only once instead of the append having to resize it.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Based on the description in commit 63ef557 this was added so that the
migrate command does not move the pause process into a separate cgroup.
It should however not disable the rejoining of the userns when the pause
process join failed. BEcause of this we end up calling migrate without a
userns and that then can fail if there are actual contianer it tries to
cleanup.
Fixes: 63ef5576ed ("command: migrate doesn't move process to cgroup")
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
There is no good reason to use logrus and os.Exit() here, other parts of
this function already return the error so do the same. The main podman
process will exit then with the normal formatted error message.
And also log an error about the last return which should never happen as
we should have exited above if the re-exec worked or errored out.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The virtiofs mount points do not actually are network mounts so we can
mount them earlier and using multi-user.target to enable them was wrong.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Currently our API docs are not working as it fails to fetch the js file
with the redoc code which renders the swagger.
The reason this fails is because we have been tracking the "next"
version and that was recently bumped to v3.0.0-rc.0 which also seem to
have moved the location to use "bundle" (no s). As such we now get a 404
as the CDN doesn't find the file. [1]
I don't get why we have been tracking next to begin with, using latest
would still work as it points but that could be moved anytime as well so
switch the URL to pull in the lastest v2 version which should be safer
against unexpected changes like that.
While it could of course also break in a minor v2 release hard coding an
exact version would mean a fair amount of churn updating this (which I
guess would not happen) so this looks like the best compromise to me.
[1] https://www.npmjs.com/package/redoc/v/3.0.0-rc.0Fixes: #27505
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The system tests can setup their own server now so just use that and
don't require the user to spawn its own.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
This got broken during the remote testing rewrite to enable tls testing
in commit feb36e4fe6. The problem is the function didn't use the proper
podman command with the remote arg so it tried to connect to the default
socket with remote testing.
And in the checkpoint file we use it before the PODMAN_CMD was defined
so fix this by moving the basic_setup before and the use the variable
PODMAN_RUNTIME instead of having to do another info call.
Also fix the debug log output to correctly log the podman commands with
all arguments on a single line.
Fixes: feb36e4fe6 ("Implement TLS API Support")
Fixes: #27266
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Make sure to rmeove the artifact but we don't need to remove the
testfiles as they are part under PODMAN_TMPDIR which gets removed by
default so don't bother with that.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
