IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
I swear I tested this, but anyways
https://github.com/projectatomic/rpm-ostree/pull/79
wasn't quite right. We need to look at /usr/etc/{passwd,group}
for previous data.
We happily noticed there was no /etc/passwd in the tree, then
proceeded to do the merge and split again, with the result
of an empty /usr/etc/passwd in the new tree.
That in turn resulted in an empty /etc/passwd in an installed system,
i.e. with no "root" user, with obvious bad consequences, namely in my
case crashing Anaconda.
(Yes, I will write a testsuite for this)
This matches recent work in OSTree to use *at() - it's faster and less
prone to error. In the case of directories which are mutable by
processes in different security domains, it's more secure too. (That's
not the case here though).
I sometimes run "rpm-ostree compose tree" directly, mainly so I can
use gdb and/or nonstandard options. In this case I don't get
version numbers injected.
That happens to trigger a bug in this code.
The checking code from #56 landed, and started triggering for me on
the `dockerroot` user. It's nice to know it works. Then the issue
is... "what now"?
It turns out in the case of `dockerroot` it's actually unused, so we
could fix this by deleting it. But in general we need to support
dynamic uids/gids/. And we can't yet take a hard dep on #49.
So this patch changes things so we take a copy of the passwd/group
data from the previous commit. Any users subsequently added in the
*new* commit will be additive.
Closes: https://github.com/projectatomic/rpm-ostree/issues/78
These files were taken from json-glib, around the era of this commit:
https://git.gnome.org/browse/json-glib/tree/build/autotools?id=2779d537492f1902d71cf648631238110b62b311
Unfortunately, this involved hacking it up a bit:
- I couldn't easily use `nobase` for the data, so I deleted that.
Test data goes in the installed-tests dir.
- Delete duplicated predeclared variables; we're using nonrecursive
make.
- Ensure we run each test in its own tmpdir
Not actually used yet; we'll need to modify these in subsequent
commits, but this is the starting point. Future commits will
therefore have a useful diff. Even if admittedly it's unlikely
they'll re-unify in the near future.
Remove redundant function _rpmostree_pull_progress().
Bumped ostree requirement to 2014.13, but this isn't quite right because
we actually need (unreleased) 2014.14. Post-release version bumps would
be useful here.
Verify uid/gid on files, directories and symlinks
Just output a msg when user/group is removed with no files
json-parsing: Add functions for strictly dealing with ints
passwd/json: Add simple scripts to convert passwd/group files to json data
docs: Check-passwd/groups and ignore-remove-users/groups JSON config. entries
It is confusing if you pull from an unversioned tree and suddenly
status loses versions. This comes at a cost of increased horizontal
space.
Closes#73
From #74
Letting GErrors pile up is wrong; either exit the function or clear the
GError (perhaps with a console warning).
In this case we tolerate missing database files (because users are free
to delete them), and build the checksum from whatever data is available.
Seems weird, but that's how I'm told it's meant to work. Bail out on
any other type of error.
Otherwise, upgraded systems which have modified /etc/group (by e.g.
adding a human user), will actually see the group drop out with bad
consequences.
It's harmless to have it in both, /etc will override /usr.
Fixes#67
We're building generic initramfs images on the server side, but dracut
has logic to pick up some things from the host, like filesystems.
In the absence of host-specific initramfs images, it needs to be up to
the generating system what kernel modules end up in the initramfs.
Provide a generic option to passthrough dracut arguments.
As a followup to renaming the "rpm" command to "db", split the "db"
subcommands into separate source files in the style of "ostree admin"
and "rpm-ostree compose".
Also create rpmostree-rpm-util.[ch] as a place for common rpm-related
functions needed by the "db" subcommands.
No intentional functional changes here, just a bunch of copy-n-paste
and minor cleanup.
Eliminates some confusion between "rpm-ostree rpm" (or "atomic rpm")
commands versus actual "rpm" commands.
The "rpm" subcommand is retained as a hidden alias for the "db"
subcommand for backward-compatibility. It is not listed in --help
output.
Fixes#22
Starting with pkg_yumdb_strdup(), the GError parameter is unused and
therefore (arguably) the GCancellable parameter is also not needed.
Remove them both, and clean up other functions that now have unused
parameters as a result.
Note that none of the callers of these functions were checking for
errors anyway.
Similar to the previous commit, but the "rpm" command needed some extra
attention.
I stopped short of splitting the subcommands into separate files (like
for "compose"), but refactored the parsing as though they were separate.
Refactor command-line parsing to better utilize GOptionContext. This
eliminates most of the manual parsing and global options are now shown
in the help output.
Some of the changes here are not strictly necessary for rpm-ostree,
but are done for consistency with ostree's command-line parsing.
The "rpm" subcommand needs some extra attention, so that's been split
into a separate commit.
=/
Originally it was "container_disabled" but the double negatives
started being awkward, I missed converting this negation.
This should really make us work again on RHEL6.
The current motivation for this is that
https://github.com/fedora-infra/fedmsg-atomic-composer
started using mock --new-chroot (which uses systemd-nspawn) to run
rpm-ostree, which in turn uses systemd-nspawn to run the post script.
Now systemd-nspawn is not really nestable (it wants to link up
journald, resolv.conf handling, etc).
First, dropping nspawn and going to raw containers fixes the nesting
problem.
Second, we don't need all the features of systemd-nspawn. We are ok
with log messages going to stdout, and we don't use networking, so no
resolv.conf is needed.
Third, this sets a bit of a stage for more sandboxing internally when
run on real systems. I already have a prototype branch which runs
librepo as an unprivileged user, that could be combined with this for
even stronger security.
Why not use systemd? Well...I'm still debating that. But the core
problem is systemd isn't a library in the C sense - to use its
sandboxing features we have to use unit files. It's harder to have a
daemon that looks like a single service from a management perspective,
but uses sandboxing internally.
"rpm-ostree compose {tree,sign} --help" acted as if no subcommand was
given. But the compose subcommands each have a good number of options
so it's particularly important to list them.
