shouldn't the current directory be the default context for "podman build"?
by Robert P. J. Day
"man podman-build" suggests that the context argument is optional:
SYNOPSIS
podman build [options] [context]
podman image build [options] [context]
...
If no context directory is specified, then Podman will assume
the current working directory as the build context, which
should contain the Containerfile.
but if i have a directory with nothing but a Containerfile, i get:
$ podman build
Error: no context directory specified, and no containerfile specified
$
OTOH, specifying context of current directory:
$ podman build .
STEP 1: FROM alpine:latest
... etc etc ...
thoughts?
rday
1 week, 4 days
Podman v5.2.2 Released
by Do Not Reply
Hi all,
Podman v5.2.2 is now available. You may view the full details at
https://github.com/containers/podman/releases/tag/v5.2.2
Release Notes:
--------------
### Bugfixes
- Fixed a bug where rootless Podman could fail to validate the runtime's volume path on systems with a symlinked `/home` ([#23515](https://github.com/containers/podman/issues/23515)).
### Misc
- Updated Buildah to v1.37.2
- Updated the containers/common library to v0.60.2
- Updated the containers/image library to v5.32.2
This message was generated by an automated system. Replies to the sender will bounce, be ignored and discarded.
4 months
Podman v5.2.1 Released
by Do Not Reply
Hi all,
Podman v5.2.1 is now available. You may view the full details at
https://github.com/containers/podman/releases/tag/v5.2.1
Release Notes:
--------------
### Bugfixes
- Fixed a bug where Podman could sometimes save an incorrect container state to the database, which could cause a number of issues including but not limited to attempting to clean up containers twice ([#21569](https://github.com/containers/podman/issues/21569)).
### Misc
- Updated Buildah to v1.37.1
- Updated the containers/common library to v0.60.1
- Updated the containers/image library to v5.32.1
This message was generated by an automated system. Replies to the sender will bounce, be ignored and discarded.
4 months, 1 week
Podman Swarm
by Daniel Perin Tavares
Hello Podmans,
Is there no implementation planned for functionality similar to 'Docker
swarm' in Podman? Swarm is so useful for small and medium infrastructures.
Daniel Perin Tavares
--
Companhia de Saneamento do Paraná
www.sanepar.com.br
<http://www.sanepar.com.br>
4 months, 1 week
new updates of using podman inside k8s
by Larry Cai
Hi:
I am used to follow podman setup inside k8s according to article
https://www.redhat.com/sysadmin/podman-inside-kubernetes
Now 3 years have passed, do we have an updated article for this?
I have a vanilla k8s env, and `privileged: true` will not be allowed soon,
so what is the easiest way to setup rootless-non-priv environment?
apiVersion: v1
kind: Pod
metadata:
name: podman
spec:
containers:
- name: podman
image: registry.access.redhat.com/ubi9/podman
command: ["sleep", "infinity"]
securityContext:
runAsUser: 1000
privileged: true
Surely `hostPath` is not allowed as well?
In the unix env, i can run `podman run -it -u podman:podman` to get correct
env
Any suggestions? (i am not security expert to understand), if I remove `
privileged: true` i got below
[podman@podman ~]$ podman system info
Error: mount
/home/podman/.local/share/containers/storage/overlay:/home/podman/.local/share/containers/storage/overlay,
flags: 0x1000: permission denied
BTW: i only need the build env, maybe `buildah` image shall be used instead
(but users are used to use podman command)
Rgs, Larry
4 months, 1 week
Why podman systemd units are deprecated
by Mehdi Haghgoo
I think podman systemd units were very useful. They allowed me to create reliable containerized services and websites on the server that would auto-restart on server reboot. However, I recently encountered podman inline help showing "deprecated" warning for systemd. Is there now a better alternative for the purpose they used to serve?
Best regards,
Mehdi
4 months, 1 week
Best practices for restricting container network access
by François-Xavier Thomas
Hello all,
I was wondering what was the current recommended way to restrict network
access of containers?
I'm trying to setup a web service via 'podman compose' ; I like most of
my web services (proxied through Nginx) to only have access to the lo
interface, and while for other services this is done via a systemd unit
setting I'm having trouble finding the equivalent for podman.
The following rules seem to work (I can also use -s/-d and specify the
whole IP range used by containers):
iptables -t filter -I NETAVARK_FORWARD -i podman2 ! -o lo,podman2 -j DROP
iptables -t filter -I NETAVARK_FORWARD -o podman2 ! -i lo,podman2 -j DROP
However, I'm not sure when the NETAVARK_FORWARD table is created (should
I even use this table?), and the podman2 interface also does not exist
before the network is created (when running 'podman compose up').
Is there a way to run these commands when the containers are brought up,
like some kind of pre-up script? Is there a better way of achieving what
I'm trying to do?
Thanks for your help,
François-Xavier
4 months, 2 weeks
Logging to journald from arbitrary processes, not just the main process
by jklaiho＠iki.fi
I've got a bunch of Quadlet-based containers with LogDriver=journald. This correctly captures the stdout/stderr of the main process of the container, and the process output is then viewable using journalctl on the host side.
We now have a need to run cron jobs in one of our containers. We're running `podman exec containername python /path/to/script.py` as the job on the host, but the stdout prints from the script are not forwarded to the logging driver.
I see two choices:
1. Run the cron scripts as short-lived containers of their own, making them the main process for the duration and presumably causing the journald driver to capture their stdout as expected. IIRC there's a way for systemd to replace cron; what would be the correct method of using Quadlet-based containers here? (We would create a separate quadlet config per script.)
2. Somehow enable arbitrary script runs with `podman exec` against a running container to interface with the journald driver. Is this possible to do?
4 months, 2 weeks
Podman v5.2.0 Released
by Do Not Reply
Hi all,
Podman v5.2.0 is now available. You may view the full details at
https://github.com/containers/podman/releases/tag/v5.2.0
Release Notes:
--------------
### Features
- Podman now supports `libkrun` as a backend for creating virtual machines on MacOS. The `libkrun` backend has the advantage of allowing GPUs to be mounted into the virtual machine to accelerate tasks. The default backend remains `applehv`.
- Quadlet now has support for `.build` files, which allows images to be built by Quadlet and then used by Quadlet containers.
- Quadlet `.container` files now support two new fields, `LogOpt` to specify container logging configuration and `StopSignal` to specify container stop signal ([#23050](https://github.com/containers/podman/issues/23050)).
- Quadlet `.container` and `.pod` files now support a new field, `NetworkAlias`, to add network aliases.
- Quadlet drop-in search paths have been expanded to include top-level type drop-ins (`container.d`, `pod.d`) and truncated unit drop-ins (`unit-.container.d`) ([#23158](https://github.com/containers/podman/issues/23158)).
- Podman now supports a new command, `podman system check`, which will identify (and, if possible, correct) corruption within local container storage.
- The `podman machine reset` command will now reset all providers available on the current operating system (e.g. ensuring that both HyperV and WSL `podman machine` VMs will be removed on Windows).
### Changes
- Podman now requires the new kernel mount API, introducing a dependency on Linux Kernel v5.2 or higher.
- Quadlet `.image` units now have a dependency on `network-online.target` ([#21873](https://github.com/containers/podman/issues/21873)).
- The `--device` option to `podman create` and `podman run` is no longer ignored when `--privileged` is also specified ([#23132](https://github.com/containers/podman/issues/23132)).
- The `podman start` and `podman stop` commands no longer print the full ID of the pod started/stopped, but instead the user's input used to specify the pod (e.g. `podman pod start b` will print `b` instead of the pod's full ID) ([#22590](https://github.com/containers/podman/issues/22590)).
- Virtual machines created by `podman machine` on Linux now use `virtiofs` instead of `9p` for mounting host filesystems. Existing mounts will be transparently changed on machine restart or recreation. This should improve performance and reliability of host mounts. This requires the installation of `virtiofsd` on the host system to function.
- Using both the `--squash` and `--layers=false` options to `podman build` at the same time is now allowed.
- Podman now passes container's stop timeout to systemd when creating cgroups, causing it to be honored when systemd stops the scope. This should prevent hangs on system shutdown due to running Podman containers.
- The `--volume-driver` option to `podman machine init` is now deprecated.
### Bugfixes
- Fixed a bug where rootless containers created with the `--sdnotify=healthy` option could panic when started ([#22651](https://github.com/containers/podman/issues/22651)).
- Fixed a bug where containers created with the `--sdnotify=healthy` option that exited quickly would sometimes return an error instead of notifying that the container was ready ([#22760](https://github.com/containers/podman/issues/22760)).
- Fixed a bug where the `podman system reset` command did not remove the containers/image blob cache ([#22825](https://github.com/containers/podman/issues/22825)).
- Fixed a bug where Podman would sometimes create a cgroup for itself even when the `--cgroups=disabled` option was specified at container creation time ([#20910](https://github.com/containers/podman/issues/20910)).
- Fixed a bug where the `/etc/hosts` file in a container was not created with a newline at the end of the file ([#22729](https://github.com/containers/podman/issues/22729)).
- Fixed a bug where the `podman start` command could sometimes panic when starting a container in the stopped state.
- Fixed a bug where the `podman system renumber` command would fail if volumes existed when using the `sqlite` database backend ([#23052](https://github.com/containers/podman/issues/23052)).
- Fixed a bug where the `podman container restore` command could not successfully restore a container in a pod.
- Fixed a bug where an error message from `podman diff` would suggest using the `--latest` option when using the remote Podman client ([#23038](https://github.com/containers/podman/issues/23038)).
- Fixed a bug where user could assign more memory to a Podman machine than existed on the host ([#18206](https://github.com/containers/podman/issues/18206)).
- Fixed a bug where the `podman events` command was rarely unable to report errors that occurred ([#23165](https://github.com/containers/podman/issues/23165)).
- Fixed a bug where containers run in systemd units would sometimes not be removed correctly on exit when using the `--cidfile` option.
- Fixed a bug where the first Podman command run after a reboot could cause hang when using transient mode ([#22984](https://github.com/containers/podman/issues/22984)).
- Fixed a bug where Podman could throw errors about a database configuration mismatch if certain paths did not exist on the host.
- Fixed a bug where the `podman run` and `podman start` commands could throw strange errors if another Podman process stopped the container at a midpoint in the process of starting ([#23246](https://github.com/containers/podman/issues/23246)).
- Fixed a bug where the `podman system service` command could leak a mount on termination.
- Fixed a bug where the Podman remote client would panic if an invalid image filter was passed to `podman images` ([#23120](https://github.com/containers/podman/issues/23120)).
- Fixed a bug where the `podman auto-update` and `podman system df` commands could fail when a container was removed while the command was running ([#23279](https://github.com/containers/podman/issues/23279)).
- Fixed a bug where the `podman machine init` command could panic when trying to decompress an empty file when preparing the VM image ([#23281](https://github.com/containers/podman/issues/23281)).
- Fixed a bug where the `podman ps --pod` and `podman pod stats` commands could sometimes fail when a pod was removed while the command was running ([#23282](https://github.com/containers/podman/issues/23282)).
- Fixed a bug where the `podman stats` and `podman pod stats` commands would sometimes exit with a `container is stopped` error when showing all containers (or pod containers, for `pod stats`) if a container stopped while the command was running ([#23334](https://github.com/containers/podman/issues/23334)).
- Fixed a bug where the output of container healthchecks was not properly logged if it did not include a final newline ([#23332](https://github.com/containers/podman/issues/23332)).
- Fixed a bug where the port forwarding firewall rules of an existing container could be be overwritten when starting a second container which forwarded the same port on the host even if the second container failed to start as the port was already bound.
- Fixed a bug where the containers created by the `podman play kube` command could sometimes not properly clean up their network stacks ([#21569](https://github.com/containers/podman/issues/21569)).
### API
- The Build API for Images now accepts a comma-separated list in the Platform query parameter, allowing a single API call to built an image for multiple architectures ([#22071](https://github.com/containers/podman/issues/22071)).
- Fixed a bug where the Remove endpoint for Volumes would return an incorrectly formatted error when called with an ambiguous volume name ([#22616](https://github.com/containers/podman/issues/22616)).
- Fixed a bug where the Stats endpoint for Containers would return an incorrectly formatted error when called on a container that did not exist ([#22612](https://github.com/containers/podman/issues/22612)).
- Fixed a bug where the Start endpoint for Pods would return a 409 error code in cases where a 500 error code should have been returned ([#22989](https://github.com/containers/podman/issues/22989)).
- Fixed a bug where the Top endpoint for Pods would return a 200 status code and then subsequently an error ([#22986](https://github.com/containers/podman/issues/22986)).
### Misc
- Podman no longer requires all parent directories of its root and runroot to be world-executable ([#23028](https://github.com/containers/podman/issues/23028)).
- Error messages from the `podman build` command when the `-f` option is given, but points to a file that does not exist, have been improved ([#22940](https://github.com/containers/podman/issues/22940)).
- The Podman windows installer is now built using WiX 5.
- Updated the gvisor-tap-vsock library to v0.7.4. This release contains a fix for a gvproxy crash on macOS when there is heavy network traffic on a fast link.
- Updated Buildah to v1.37.0
- Updated the containers/image library to v5.32.0
- Updated the containers/storage library to v1.55.0
- Updated the containers/common library to v0.60.0
This message was generated by an automated system. Replies to the sender will bounce, be ignored and discarded.
4 months, 3 weeks
