This was previously tied to an older commit rev and not an actual release
version; and that old commit was causing connectivity issues due to improper
SSL support.
Since systemd version 230, it is required to have a machine-id file
prior to the startup of the container. If the file is empty, a transient
machine ID is generated by systemd-nspawn.
See systemd/systemd#3014 for more details on the matter.
This unbreaks all of the containers-* NixOS tests.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
Cc: @edolstra
Closes: #15808
If /tmp and /nix are in different filesystems, this causes the `find`s
in the fixup phase to fail because of a stale file handle:
find: cannot get current directory: No such file or directory
[Breakage introduced in commit 5c4e00b6b797c444c0a0ca434c3b6f
("gst_all_1: 1.6.1 -> 1.8.0 (#14628)").]
The build is failing due to a Wayland test/example program
(<gst-plugins-bad>/tests/examples/waylandsink) being built with a gtk3
input without Wayland support (or detection is failing):
main.c:28:2: error: #error "Wayland is not supported in GTK+"
Fix it by explicitly disabling wayland; pass --enable-wayland=no to
configure.
This reverts commit 83406bc171ca2b385c49838e6ed5e89007a875b5, because
it broke the build.
x2goclient requires to be built with its top-level (hand coded) Makefile
(in accordance with upstream documentation). Invoking qmake directly on
the .pro file, without specifying a separate build tree, will overwrite
the Makefile and break the build.
For instance, there are no install rules in the .pro file. That exists
only in the Makefile.
This is the original pull request plus some commits from me to bring all
channels to the latest versions, because the fixed security
vulnerabilites might not be fixed in the dev version we had before.
I've tested the whole changeset on my Hydra at:
https://headcounter.org/hydra/eval/322006
Thanks to @srp for the initial commit and thus implicitly also for the
security notice.
Cc: @abbradar
Refers to non-existent files; see e.g.,
https://hydra.nixos.org/build/36359717/nixlog/1/raw
Likely a copy-paste error that has gone unnoticed because paxmark didn't
do anything, but breaks after 6648b04381b8fefb704824f5db898813f22dafbb
With this update we need to rebase the nix_plugin_paths patch, which was
done by @srp and I took it from his comment at:
https://github.com/NixOS/nixpkgs/pull/15762#issuecomment-222230677
Other than that, using libjpeg from nixpkgs fails to link:
https://headcounter.org/hydra/build/1114273
Rather than just using versionAtLeast to check for >= version 52, we're
matching on the explicit version number. That way we can make sure that
we (try to) build with system libjpeg again so we can keep it out of the
overall Chromium build time.
Built and tested using the VM tests on my Hydra at:
https://headcounter.org/hydra/eval/322006
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
We're already on version 52, so there really is no need to keep all
those conditionals and old patches anymore.
Tested dropping the unconditional build_fixes_46.patch via the Chromium
VM tests.
Signed-off-by: aszlig <aszlig@redmoonstudios.org>
