The CONTAINERS_CONF environment variable can be used to override the
configuration file, which is useful for testing. However, at the moment
this variable is not propagated to conmon. That means in particular, that
conmon can't propagate it back to podman when invoking its --exit-command.
The mismatch in configuration between the starting and cleaning up podman
instances can cause a variety of errors.
This patch also adds two related test cases. One checks explicitly that
the correct CONTAINERS_CONF value appears in conmon's environment. The
other checks for a possible specific impact of this bug: if we use a
nonstandard name for the runtime (even if its path is just a regular crun),
then the podman container cleanup invoked at container exit will fail.
That has the effect of meaning that a container started with -d --rm won't
be correctly removed once complete.
Fixes#12917
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Waiting on an initialized sync.WaitGroup returns immediately.
Hence, move the goroutine to wait and close *after* reading
the logs.
Fixes: #12904
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
It would be easier to diagnose OCI runtime errors if the error actually
had the name of the OCI runtime that produced the error.
[NO NEW TESTS NEEDED]
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
We should not modify the XDG_RUNTIME_DIR env value during runtime of
libpod, this can cause hard to find bugs. Only set it for the OCI
runtime, this matches the other commands such as start, stop, kill...
[NO NEW TESTS NEEDED]
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
if the SELinux label could not be restored correctly, leave the OS
thread locked so that it is terminated once it returns to the threads
pool.
[NO NEW TESTS NEEDED] the failure is hard to reproduce
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
This should fix the SELinux issue we are seeing with talking to
/run/systemd/private.
Fixes: https://github.com/containers/podman/issues/12362
Also unset the XDG_RUNTIME_DIR if set, since we don't know when running
as a service if this will cause issue.s
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
failed to send a signal to the container's PID1, but ignored the
results of that update. That's generally bad practice, since even
if we can't directly take action on an error, we should still
make an effort to report it for debugging purposes. I used Infof
instead of something more serious to avoid duplicate reporting to
the user if something has gone seriously wrong.
[NO NEW TESTS NEEDED] this is just adding additional error reporting.
Signed-off-by: Matthew Heon <mheon@redhat.com>
`crun status ctrid` outputs `No such file or directory` when container
is not there so podman much ack it.
[NO NEW TESTS NEEDED]
Signed-off-by: Aditya Rajan <arajan@redhat.com>
While trying to kill a container with a `signal` we cant do anything if
container is already dead so `exit` gracefully instead of trying to
delete container again. Get container status from runtime.
[ NO NEW TESTS NEEDED ]
Signed-off-by: Aditya Rajan <arajan@redhat.com>
There is a problem with creating and storing the exit command when the
container was created. It only contains the options the container was
created with but NOT the options the container is started with. One
example would be a CNI network config. If I start a container once, then
change the cni config dir with `--cni-config-dir` ans start it a second
time it will start successfully. However the exit command still contains
the wrong `--cni-config-dir` because it was not updated.
To fix this we do not want to store the exit command at all. Instead we
create it every time the conmon process for the container is startet.
This guarantees us that the container cleanup process is startet with
the correct settings.
[NO NEW TESTS NEEDED]
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
CRIU supports checkpoint/restore of file locks. This feature is
required to checkpoint/restore containers running applications
such as MySQL.
Signed-off-by: Radostin Stoyanov <radostin@redhat.com>
This adds the parameter '--print-stats' to 'podman container restore'.
With '--print-stats' Podman will measure how long Podman itself, the OCI
runtime and CRIU requires to restore a checkpoint and print out these
information. CRIU already creates process restore statistics which are
just read in addition to the added measurements. In contrast to just
printing out the ID of the restored container, Podman will now print
out JSON:
# podman container restore --latest --print-stats
{
"podman_restore_duration": 305871,
"container_statistics": [
{
"Id": "47b02e1d474b5d5fe917825e91ac653efa757c91e5a81a368d771a78f6b5ed20",
"runtime_restore_duration": 140614,
"criu_statistics": {
"forking_time": 5,
"restore_time": 67672,
"pages_restored": 14
}
}
]
}
The output contains 'podman_restore_duration' which contains the
number of microseconds Podman required to restore the checkpoint. The
output also includes 'runtime_restore_duration' which is the time
the runtime needed to restore that specific container. Each container
also includes 'criu_statistics' which displays the timing information
collected by CRIU.
Signed-off-by: Adrian Reber <areber@redhat.com>
This adds the parameter '--print-stats' to 'podman container checkpoint'.
With '--print-stats' Podman will measure how long Podman itself, the OCI
runtime and CRIU requires to create a checkpoint and print out these
information. CRIU already creates checkpointing statistics which are
just read in addition to the added measurements. In contrast to just
printing out the ID of the checkpointed container, Podman will now print
out JSON:
# podman container checkpoint --latest --print-stats
{
"podman_checkpoint_duration": 360749,
"container_statistics": [
{
"Id": "25244244bf2efbef30fb6857ddea8cb2e5489f07eb6659e20dda117f0c466808",
"runtime_checkpoint_duration": 177222,
"criu_statistics": {
"freezing_time": 100657,
"frozen_time": 60700,
"memdump_time": 8162,
"memwrite_time": 4224,
"pages_scanned": 20561,
"pages_written": 2129
}
}
]
}
The output contains 'podman_checkpoint_duration' which contains the
number of microseconds Podman required to create the checkpoint. The
output also includes 'runtime_checkpoint_duration' which is the time
the runtime needed to checkpoint that specific container. Each container
also includes 'criu_statistics' which displays the timing information
collected by CRIU.
Signed-off-by: Adrian Reber <areber@redhat.com>
we are having a hard time figuring out a failure in the CI:
https://github.com/containers/podman/issues/11191
Rename the sub-cgroup created here, so we can be certain the error is
caused by this part.
[NO NEW TESTS NEEDED] we need this for the CI.
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
Note: the Warning message will not come to podman-remote.
It would be difficult to plumb, and not really worth the effort.
Fixes: https://github.com/containers/podman/issues/11854
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
There is a race where `conn.Close()` was called before `conn.CloseWrite()`.
In this case `CloseWrite` will fail and an useless error is printed. To
fix this we move the the `CloseWrite()` call to the same goroutine to
remove the race. This ensures that `CloseWrite()` is called before
`Close()` and never afterwards.
Also fixed podman-remote run where the STDIN was never was closed.
This is causing flakes in CI testing.
[NO TESTS NEEDED]
Fixes#11856
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The backend for `ps --sync` has been nonfunctional for a long
while now - probably since v2.0. It's questionable how useful the
flag is in modern Podman (the original case it was intended to
catch, Conmon gone via SIGKILL, should be handled now via pinging
the process with a signal to ensure it's still alive) but having
the ability to force a refresh of container state from the OCI
runtime is still useful.
Signed-off-by: Matthew Heon <mheon@redhat.com>
Access the container's config field directly inside of libpod instead of
calling `Config()` which in turn creates expensive JSON deep copies.
Accessing the field directly drops memory consumption of a simple
`podman run --rm busybox true` from 1245kB to 410kB.
[NO TESTS NEEDED]
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
it allows to pass the current std streams down to the container.
conmon support: https://github.com/containers/conmon/pull/289
[NO TESTS NEEDED] it needs a new conmon.
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
For rootful users ports are forwarded via iptables. To make sure no
other process tries to use them, libpod will bind the ports and pass the
fds to conmon. There seems to be race when a container is restarted
because libpod tries to bind the port before the conmon process exited.
The problem only hapens with the podman service because it keeps the
connection open. Once we have the fd and passed it to conmon the
podman service should close the connection.
To verify run `sudo ss -tulpn` and check that only the conmon process
keeps the port open. Previously you would also see the podman server
process listed.
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
The conmon buffer size is 8192, however the attach socket needs two extra
bytes. The first byte of each message will be the STREAM type. The last
byte is a null byte. So when we want to read 8192 message bytes we need
to read 8193 bytes since the first one is special.
check 1ef246896b/src/ctr_stdio.c (L101-L107)
This problem can be seen in podman-remote run/exec when it prints output
with 8192 or more bytes. The output will miss the 8192 byte.
Fixes#11496
Signed-off-by: Paul Holzinger <pholzing@redhat.com>
Make sure that Podman passes the LISTEN_* environment into containers.
Similar to runc, LISTEN_PID is set to 1.
Also remove conditionally passing the LISTEN_FDS as extra files.
The condition was wrong (inverted) and introduced to fix#3572 which
related to running under varlink which has been dropped entirely
with Podman 3.0. Note that the NOTIFY_SOCKET and LISTEN_* variables
are cleared when running `system service`.
Fixes: #10443
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
In libpod/logs.LogLine.Write(), don't write a newline to stdout/stderr
when the log message is only part of a line.
In libpod.ConmonOCIRuntime.HTTPAttach(), don't send a newline over the
HTTP connection when the log message is only part of a line.
In pkg/api/handlers/compat.LogsFromContainer(), don't send a newline
over the HTTP connection when the log message is only part of a line,
and don't make doing so conditional on whether or not the client used
the docker or podman endpoint.
In pkg/domain/infra/tunnel.ContainerEngine.ContainerLogs(), don't add
our own newline to log messages, since they already come through from
the server when they need to.
Signed-off-by: Nalin Dahyabhai <nalin@redhat.com>
This leverages conmon's ability to proxy the SD-NOTIFY socket.
This prevents locking caused by OCI runtime blocking, waiting for
SD-NOTIFY messages, and instead passes the messages directly up
to the host.
NOTE: Also re-enable the auto-update tests which has been disabled due
to flakiness. With this change, Podman properly integrates into
systemd.
Fixes: #7316
Signed-off-by: Joseph Gooch <mrwizard@dok.org>
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
This adds support to checkpoint containers out of pods and restore
container into pods.
It is only possible to restore a container into a pod if it has been
checkpointed out of pod. It is also not possible to restore a non pod
container into a pod.
The main reason this does not work is the PID namespace. If a non pod
container is being restored in a pod with a shared PID namespace, at
least one process in the restored container uses PID 1 which is already
in use by the infrastructure container. If someone tries to restore
container from a pod with a shared PID namespace without a shared PID
namespace it will also fail because the resulting PID namespace will not
have a PID 1.
Signed-off-by: Adrian Reber <areber@redhat.com>
Unfortunately --pre-checkpointing never worked as intended and recent
changes to runc have shown that it is broken.
To create a pre-checkpoint CRIU expects the paths between the
pre-checkpoints to be a relative path. If having a previous checkpoint
it needs the be referenced like this: --prev-images-dir ../parent
Unfortunately Podman was giving runc (and CRIU) an absolute path.
Unfortunately, again, until March 2021 CRIU silently ignored if
the path was not relative and switch back to normal checkpointing.
This has been now fixed in CRIU and runc and running pre-checkpoint
with the latest runc fails, because runc already sees that the path is
absolute and returns an error.
This commit fixes this by giving runc a relative path.
This commit also fixes a second pre-checkpointing error which was just
recently introduced.
So summarizing: pre-checkpointing never worked correctly because CRIU
ignored wrong parameters and recent changes broke it even more.
Now both errors should be fixed.
[NO TESTS NEEDED]
Signed-off-by: Adrian Reber <areber@redhat.com>
Signed-off-by: Adrian Reber <adrian@lisas.de>
Commit 728b73d7c418 introduced a regression. Containers created with a
previous version do no longer start successfully. The problem is that
the PidFile in the container config is empty for those containers. If
the PidFile is empty we have to set it to the previous default.
[NO TESTS NEEDED] We should investigate why the system upgrade test did
not caught this.
Fixes#10274
Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
This option allows users to specify the maximum amount of time to run
before conmon sends the kill signal to the container.
Fixes: https://github.com/containers/podman/issues/6412
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Currently the debug line shows every runtime up until it finds
the correct one, confusing users on which runtime it is using.
Also move missing OCI runtime from containers/conf down to Debug level
and improved the debug message, to not report error.
[NO TESTS NEEDED] Since this is just debug.
Triggered by https://github.com/containers/podman/issues/4854
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
One of the side-effects of the `--userns=keep-id` command is
switching the default user of the container to the UID of the
user running Podman (though this can still be overridden by the
`--user` flag). However, it did this by setting the UID and GID
in the OCI spec, and not by informing Libpod of its intention to
switch users via the `WithUser()` option. Because of this, a lot
of the code that should have triggered when the container ran
with a non-root user was not triggering. In the case of the issue
that this fixed, the code to remove capabilities from non-root
users was not triggering. Adjust the keep-id code to properly
inform Libpod of our intention to use a non-root user to fix
this.
Also, fix an annoying race around short-running exec sessions
where Podman would always print a warning that the exec session
had already stopped.
Fixes#9919
Signed-off-by: Matthew Heon <matthew.heon@pm.me>
when --privileged is used, make sure to not request more capabilities
than currently available in the current context.
[NO TESTS NEEDED] since it fixes existing tests.
Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
Fixes: https://github.com/containers/podman/issues/9582
This PR also adds tests to make sure SELinux labels match the runtime,
or if init is specified works with the correct label.
Add tests for selinux kvm/init labels
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
To be able to reuse common checkpoint/restore functions this commit
moves code to pkg/checkpoint/crutils.
This commit has not functional changes. It only moves code around.
[NO TESTS NEEDED] - only moving code around
Signed-off-by: Adrian Reber <areber@redhat.com>
prune a dependency that was only being used for a simple struct. Should
correct checksum issue on tarballs
[NO TESTS NEEDED]
Fixes: #9355
Signed-off-by: baude <bbaude@redhat.com>
We missed bumping the go module, so let's do it now :)
* Automated go code with github.com/sirkon/go-imports-rename
* Manually via `vgrep podman/v2` the rest
Signed-off-by: Valentin Rothberg <rothberg@redhat.com>
`staticcheck` is a golang code analysis tool. https://staticcheck.io/
This commit fixes a lot of problems found in our code. Common problems are:
- unnecessary use of fmt.Sprintf
- duplicated imports with different names
- unnecessary check that a key exists before a delete call
There are still a lot of reported problems in the test files but I have
not looked at those.
Signed-off-by: Paul Holzinger <paul.holzinger@web.de>
