https://gitlab.com/gitlab-org/gitaly/-/commit/e7f8fe4f6759669c24b2fd216beaab4a477c5b3d
changed the binary name to always end with its module version. This
makes sure gitaly's internal version references are set to the package
version and renames the binary, postfixing it with the package
version.
(cherry picked from commit d14e9188d1cd88f1b530bf860638c2de27486a6d)
Some plugin repos already have a `gems` directory. This lets the
packager choose whether it should be kept and the nix packaged ruby
gems should be copied into it or if it should be removed in favor of
our ruby gems.
(cherry picked from commit 04e6b03fa91603c7f1961cfcdcf5880c91fe6b05)
Let the update.py script handle the initial, repetitive task of
packaging new plugins. With this in place, the plugin only needs to be
added to the list in `update-plugins` and most of the work will be
done automatically when the script is run. Metadata still needs to be
filled in manually and some packages may of course require additional
work/patching.
(cherry picked from commit f8096460bd15d4f13a01cfddf0a30798921fdb42)
Change the path to the auto generated plugin assets, which defaults to
the plugin's directory and isn't writable at the time of asset
generation.
(cherry picked from commit 443b318ee9c614d480a2ecb0120b52806d6fbb3b)
Perform the tests on the package that the `tests` attribute is a child
of, i.e. if `discourseAllPlugins.tests` is built, the tests will run
with the `discourseAllPlugins` package, not the `discourse` package as
previously.
(cherry picked from commit 6fd5a40ccaf0b4da1362803a387bf46d381dd66a)
Maybe bundix doesn't handle them properly? They cause runtime issues
and don't seem necessary when the binary gems are built from scratch
anyway.
(cherry picked from commit a2dbc3af1d50edb8d646817e7abe799bcf369f20)
Some discourse plugins have Ruby dependencies and require a
specialized builder. This introduces a generic builder that can be
used whether the plugin has Ruby dependencies or not. It also adds a
set of pre-packaged plugins available through `discourse.plugins` and
provides an easy way to add more.
(cherry picked from commit 7671b90919af8271bdf7e07bb86d63216a2145d6)
For plugins to work properly, their assets need to be precompiled
along with the rest of Discourse's assets. This means we need to build
new packages when the list of plugins change.
(cherry picked from commit 9af3672f4faaafba0ce0129a87fc7925c14eeb61)
This makes the service fail when upgrading the package, so let's
properly restart it instead.
(cherry picked from commit b4c069b1476a92a540e906ef95cd7fb380d29c63)
62733b37b4 broke evaluation in all
places `pkgs.mysql` was used. Fix this by changing all occurrences to
`pkgs.mariadb`.
(cherry picked from commit 59e0120aa5c1241d48048afa615e25c65d7e366d)
Assert that the PostgreSQL version being deployed is the one used
upstream. Allow the user to override this assertion, since it's not
always possible or preferable to use the recommended one.
(cherry picked from commit 544adbfcab2e92c2fe5774cae67f2edf165eb97e)
Instead of requiring the user to bundle the certificate and private
key into a single file, provide separate options for them. This is
more in line with most other modules.
`install` copies the files before setting their mode, so there could
be a breif window where the secrets are readable by other users
without a strict umask.
Feeding `psql` the password on the command line leaks it through the
`psql` process' `/proc/<pid>/cmdline` file. Using `echo` to put the
command in a file and then feeding `psql` the file should work around
this, since `echo` is a bash builtin and thus shouldn't spawn a new
process.
Using `replace-literal` to insert secrets leaks the secrets through
the `replace-literal` process' `/proc/<pid>/cmdline`
file. `replace-secret` solves this by reading the secret straight from
the file instead, which also simplifies the code a bit.
Using `replace-literal` to insert secrets leaks the secrets through
the `replace-literal` process' `/proc/<pid>/cmdline`
file. `replace-secret` solves this by reading the secret straight from
the file instead.
Using `replace-literal` to insert secrets leaks the secrets through
the `replace-literal` process' `/proc/<pid>/cmdline`
file. `replace-secret` solves this by reading the secret straight from
the file instead, which also simplifies the code a bit.
Using `replace-literal` to insert secrets leaks the secrets through
the `replace-literal` process' `/proc/<pid>/cmdline`
file. `replace-secret` solves this by reading the secret straight from
the file instead, which also simplifies the code a bit.
Using `replace-literal` to insert secrets leaks the secrets through
the `replace-literal` process' `/proc/<pid>/cmdline`
file. `replace-secret` solves this by reading the secret straight from
the file instead, which also simplifies the code a bit.
Add a small utility script which securely replaces secrets in
files. Doing this with `sed`, `replace-literal` or similar utilities
leaks the secrets through the spawned process' `/proc/<pid>/cmdline` file.
If `/run/opengl-driver/lib` is in `LD_LIBRARY_PATH`, bwrap will try to
mount it into the already bind mounted `/run/opengl-driver`. To
prevent this, we instead specify the subpaths of `/run/opengl-driver`
we're interested in.
Also, use the correct path for i686 builds.
This reverts commit d9e18f4e7f.
This change is broken, since it doesn't configure the proper database
username in keycloak when provisioning a local database with a custom
username. Its intended behavior is also potentially confusing and
dangerous, so rather than fixing it, let's revert to the old one.
This reverts commit d6e0d38b84.
We need shorter secrets to continue working, since the earlier
recommendation was too short and there's no way to rotate the them.
The upstream recommended minimum length for db_key_base is 30 bytes,
which our option descriptions repeated. Recently, however, upstream
has, in many places, moved to using aes-256-gcm, which requires a key
of exactly 32 bytes. To allow for shorter keys, the upstream code pads
the key in some places. However, in many others, it just truncates the
key if it's too long, leaving it too short if it was to begin
with. This adds a patch that fixes this and updates the descriptions
to recommend a key of at least 32 characters.
See https://gitlab.com/gitlab-org/gitlab/-/merge_requests/53602
Bash doesn't handle subshell errors properly if the result is used as
input to a command. To cause the services to fail when the files can't
be read, we need to assign the value to a variable, then export it
separately.
Follow the interface of the `nginx` package and expose the enabled
modules. This is used in the `nginx` module to enable the
`additionalModules` option.
This should reduce memory fragmentation drastically and is recommended
by both the Puma and the Sidekiq author. It's also the default value
for Ruby deployments on Heroku.
Kim Lindberger