Works for me. Thank you. Fixes BZ#2283290

cc https://github.com/containers/bootc/issues/509#issuecomment-2156630865

OK yes, as far as I can tell this update changed a lot of things in containers.conf

$ curl -L https://kojipkgs.fedoraproject.org//packages/containers-common/0.59.1/1.fc40/noarch/containers-common-0.59.1-1.fc40.noarch.rpm | rpm2cpio | cpio -div
$ curl -L https://kojipkgs.fedoraproject.org//packages/containers-common/0.58.0/2.fc40/noarch/containers-common-0.58.0-2.fc40.noarch.rpm |rpm2cpio | cpio -div
$ diff -uNr old new

turns up:

diff -uNr old/usr/share/containers/containers.conf new/usr/share/containers/containers.conf
--- old/usr/share/containers/containers.conf    2024-06-11 11:20:58.530160190 -0400
+++ new/usr/share/containers/containers.conf    2024-06-11 11:21:06.673289954 -0400
@@ -191,7 +191,8 @@

 # Logging driver for the container. Available options: k8s-file and journald.
 #
-#log_driver = "journald"
+#log_driver = "k8s-file"
+log_driver = "journald"

 # Maximum size allowed for the container log file. Negative numbers indicate
 # that no size limit is imposed. If positive, it must be >= 8192 to match or
@@ -350,9 +351,9 @@

 # The firewall driver to be used by netavark.
 # The default is empty which means netavark will pick one accordingly. Current supported
-# drivers are "iptables", "none" (no firewall rules will be created) and "firewalld" (firewalld is
-# experimental at the moment and not recommend outside of testing). In the future we are
-# planning to add support for a "nftables" driver.
+# drivers are "iptables", "nftables", "none" (no firewall rules will be created) and "firewalld" (firewalld is
+# experimental at the moment and not recommend outside of testing).
+#
 #firewall_driver = ""


@@ -440,6 +441,7 @@
 # to v2s2.
 #
 #compression_format = "gzip"
+compression_format = "zstd:chunked"

 # The compression level to use when pushing an image.
 # Valid options depend on the compression format used.
@@ -529,6 +531,15 @@
 # with detailed information about the container.
 #events_container_create_inspect_data = false

+# Whenever Podman should log healthcheck events.
+# With many running healthcheck on short interval Podman will spam the event
+# log a lot as it generates a event for each single healthcheck run. Because
+# this event is optional and only useful to external consumers that may want
+# to know when a healthcheck is run or failed allow users to turn it off by
+# setting it to false. The default is true.
+#
+#healthcheck_events = true
+
 # A is a list of directories which are used to search for helper binaries.
 #
 #helper_binaries_dir = [
@@ -544,6 +555,12 @@
 #  "/usr/share/containers/oci/hooks.d",
 #]

+# Directories to scan for CDI Spec files.
+#
+#cdi_spec_dirs = [
+#  "/etc/cdi",
+#]
+
 # Manifest Type (oci, v2s2, or v2s1) to use when pulling, pushing, building
 # container images. By default image pulled and pushed match the format of the
 # source image. Building/committing defaults to OCI.
@@ -672,6 +689,7 @@
 # Default OCI runtime
 #
 #runtime = "crun"
+runtime = "crun"

 # List of the OCI runtimes that support --format=json. When json is supported
 # engine will use it for reporting nicer errors.
@@ -744,9 +762,6 @@
 # A value of 0 is treated as no timeout.
 #volume_plugin_timeout = 5

-# Default timeout in seconds for podmansh logins.
-#podmansh_timeout = 30
-
 # Paths to look for a valid OCI runtime (crun, runc, kata, runsc, krun, etc)
 [engine.runtimes]
 #crun = [
@@ -874,3 +889,14 @@
 #
 # map of existing farms
 #[farms.list]
+
+[podmansh]
+# Shell to spawn in container. Default: /bin/sh.
+#shell = "/bin/sh"
+#
+# Name of the container the podmansh user should join.
+#container = "podmansh"
+#
+# Default timeout in seconds for podmansh logins.
+# Favored over the deprecated "podmansh_timeout" field.
+#timeout = 30

I think switching the default compression format is not something that probably should happen in the middle of a stable release.

What's baffling me at the moment is: where the heck does that containers.conf come from? I don't see that change in teh upstream https://github.com/containers/common/blob/main/pkg/config/containers.conf

OK wow, there's apparently this thing https://github.com/containers/common/commit/5c5b1120aaa2970d1b6c961ba2d48996fb240c6a#diff-a34e8a216364f83f74176a4c9f68c05f152789fb5bb7b00999b301181e362f77 that patches the default config files...

OK I think I found it https://github.com/containers/common/commit/5c5b1120aaa2970d1b6c961ba2d48996fb240c6a#r143013047

So we clearly need https://github.com/containers/common/commit/d3ca62d55d5b4797860c37888db040954e5aa7e0 in the stable release ASAP.

→ FEDORA-2024-ebe5c816a1

This update also moved use-overlayfs to a suggests, which means that it got removed from Fedora Atomic Desktops: https://pagure.io/workstation-ostree-config/pull-request/526