When Podman starts, it checks a number of critical runtime paths
against stored values in the database to make sure that existing
containers are not broken by a configuration change. We recently
made some changes to this logic to make our handling of the some
options more sane (StaticDir in particular was set based on other
passed options in a way that was not particularly sane) which has
made the logic more sensitive to paths with symlinks. As a simple
fix, handle symlinks properly in our DB vs runtime comparisons.
The BoltDB bits are uglier because very, very old Podman versions
sometimes did not stuff a proper value in the database and
instead used the empty string. SQLite is new enough that we don't
have to worry about such things.
Fixes#20872
Signed-off-by: Matt Heon <mheon@redhat.com>
Right now, we always use a private UTS namespace on FreeBSD. This should
be made optional but implementing that cleanly needs a FreeBSD extension
to the OCI runtime config. The process for that is starting
(https://github.com/opencontainers/tob/pull/133) but in the meantime,
assume that the UTS namespace is private on FreeBSD.
This moves the Linux-specific namespace logic to
container_internal_linux.go and adds a FreeBSD stub.
[NO NEW TESTS NEEDED]
Signed-off-by: Doug Rabson <dfr@rabson.org>
When a new API call is added to the bindings we should guard it based on
the version and throw a useful error. Right now an old server that does
not implement a given endpoint would throw a "NOT FOUND" error which is
not good for callers.
Instead implement a custom error type to give a usefule error instead.
This allows bindings users to call errors.As() to know if they call and
to old version.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Commit f48a706abc added a new API endpoint to remove exec session
correctly. And the bindings try to call that endpoint for exec every
time. Now since client and server must not be the same version this
causes a problem if a new 4.8 client calls an older 4.7 server as it has
no idea about such endpoint and throws an ugly error. This is a common
scenario for podman machine setups.
The client does know the server version so it should make sure to not
call such endpoint if the server is older than 4.8.
I added a exec test to the machine tests as this can be reproduced with
podman machine as at the moment at least the VM image does not contain
podman 4.8. And it should at least make sure podman exec keeps working
for podman machine without regressions.
Fixes#20821
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
If a transaction is started it must either be committed or rolled back.
The function uses defer to call `tx.Rollback()` if there is an error
returned. However it also called `tx.Commit()` and afterwards further
errors can be returned which means it tries to roll back a already
committed transaction which cannot work.
This fix is to make sure tx.Commit() is the last call in that function.
see https://github.com/containers/podman/issues/20731
[NO NEW TESTS NEEDED]
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
We have to Commit() the transaction. Note this is only in a rare pod
remove code path and very unlikely to ever be used.
[NO NEW TESTS NEEDED]
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Only one process can write to the sqlite db at the same time, if another
process tries to use it at that time it fails and a database is locked
error is returned. If this happens sqlite should keep retrying until it
can write. To do that we can just set the _busy_timeout option. A 100s
timeout should be enough even on slower systems but not to much in case
there is a deadlock so it still returns in a reasonable time.
[NO NEW TESTS NEEDED] I think we strongly need to consider some form of
parallel stress testing to catch bugs like this.
Fixes#20809
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Fixed a bug where `podman machine rm -f` would cause a deadlock when
running with WSL.
The deadlock is caused by the Remove() function calling the Stop()
function after Remove() locks the VM. Stop() also has a lock call, which
fails and deadlocks because Remove() already claimed lock. Fix this by
moving the stop call before the lock
[NO NEW TESTS NEEDED]
Signed-off-by: Ashley Cui <acui@redhat.com>
Two newly-added tests, fail in gating:
- system connection: difference in how sockets are set up
between CI and gating
- ulimit: gating seems to run with ulimit -c -H 0. Check, and
skip if ulimit is less than what we need
Signed-off-by: Ed Santiago <santiago@redhat.com>
Currently if user specifies podman kube play --replace, the
pod is removed on the client side, not the server side. If
the API is called with replace=true, the pod was not being removed
and this called the API to fail. This PR removes the pod if it
exists and the caller specifies replace=true.
Fixes: https://github.com/containers/podman/discussions/20705
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
The 5.0 release, which is coming up fast, will make a number of
deprecations and removals. We can start giving folks a heads up
on them now, so there is no confusion when 5.0 drops. We will
also write some blogs about the transition over the coming weeks
(after we have a better idea of what the final release will look
like).
Signed-off-by: Matt Heon <mheon@redhat.com>
As the title says. Of most importance, this contains a bump of
BuildKit to v0.12.3 which should reduce the bloat over the
prior version of BuildKit.
[NO NEW TESTS NEEDED]
Signed-off-by: TomSweeneyRedHat <tsweeney@redhat.com>
If gvproxy or vfkit exit we can error right away, so while we wait for
the socket to get ready we also keep checking the process status with
wait4() and WNOHANG so it does not block forever.
This is completely untested as I do not have acces to apple machine.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
<MH: Added no new tests needed to pass CI>
[NO NEW TESTS NEEDED]
Signed-off-by: Matt Heon <mheon@redhat.com>
- We can assume that cgroups v2 and rootless overlayfs are the
default everywhere.
- Remove RHEL7-only instructions
- add clear '$' and '#' prompts to rootless and root commands
- other minor consistency cleanups
Ref: #20669
Signed-off-by: Ed Santiago <santiago@redhat.com>
Bump Buildah to v1.33.1 to get a CVE fix for Buildkit.
I thought it was also going to drag in the test fix as
mentioned in #20709, but I'm not seeing that here.
[NO NEW TESTS NEEDED]
Signed-off-by: TomSweeneyRedHat <tsweeney@redhat.com>
As the title says. This is the last dance step in preparation
for Podman v4.8.
[NO NEW TESTS NEEDED]
Signed-off-by: TomSweeneyRedHat <tsweeney@redhat.com>
florent found a bug where he used "applehv" as a machine name. it turns out when we use a vmtype name, esp. the active type, it really messes up directory structures for configuration and images alike.
Signed-off-by: Brent Baude <bbaude@redhat.com>
This mostly just inherits the c/common/pkg/auth implementation,
except that AuthFilePath and DockerCompatAuthFilePath can not be set
simultaneously, so don't unnecessarily explicitly set AuthFilePath.
c/common already handles that.
Signed-off-by: Miloslav Trmač <mitr@redhat.com>
