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.
We had duplicated code across the cmdline entrypoints for transaction
processing; things like "print pkg diff only if !opt_reboot".
This doesn't dedup all of them - there are some corner cases around
the preview logic in `upgrade`, and `initramfs` also need special
handling. I'll likely enhance this further down the line for that.
But one reason I'm doing this now is prep for:
[rpm-ostree cancel](https://github.com/projectatomic/rpm-ostree/pull/1019)
Basically, I want to add a `-B/--background` option we honor consistently, and
that'd be a lot easier if we have a combined "start/monitor txn" with the
post-txn option processing in one place.
Closes: #1034
Approved by: jlebon
There are (somewhat hard to avoid) race conditions for the client
to find the active transaction and connect to it. This approach
adds a retry loop, and teaches the status builtin to use it.
The status code becomes a lot less ugly. Prep for:
[rpm-ostree cancel](https://github.com/projectatomic/rpm-ostree/pull/1019)
Closes: #1033
Approved by: peterbaouoft
Now that we have the ability to both track clients and our active transaction,
and the `RegisterClient` call acts "atomically", let's start doing exit-on-idle
and return the RAM to the people.
Closes: #606
Approved by: jlebon
This closes a race condition for having the daemon idle exit. After
the daemon has released its bus name, the dbus-daemon will no longer
allow messages through that targeted its unique name.
Since the intention of the `RegisterClient` method is to be the "knock on the
door", fix this by directly sending a message to the well-known name.
Second, we need to handle the case where the daemon exits without
replying; @jlebon added a `sleep(10)` invocation after the daemon
mainloop quit but before we `ReleaseName`, and I verified these two
things combine to fix that case.
Closes: #606
Approved by: jlebon
Prep for implementing `rpm-ostree cancel`, but this works with the way we handle
`Ctrl-C` interactively on a client as well. Being able to cancel a script
execution is quite nice; some of them are expensive, and having one loop forever
has been known to happen.
Closes: #1025
Approved by: jlebon
There's a lot of paths in the core related to SELinux policy changes and
relabeling packages. We currently have no test coverage for them. We add
support in the test libraries here to build such packages.
We also add a test that checks both that we correctly relabel RPMs when
the policy changes and that we handle layered packages that install
SELinux packages properly.
Closes: #999
Approved by: cgwalters
It's possible that a layered package installed an SELinux package which
can change the labeling of some files. The problem is that imported
packages will not have used the updated policy and thus may have some of
its paths mislabeled. We fix this here by always loading the SELinux
policy from the tmprootfs and using that for the final commit.
Closes: #999
Approved by: cgwalters
Nothing actually uses it, we've been all fd-relative for a while. Just noticed
this while looking at the compose code for further work.
Closes: #1030
Approved by: jlebon
Switch `ex container` to `OSTREE_REPO_MODE_BARE_USER_ONLY`; this is a good
match, for the same reasons as flatpak. We don't want suid binaries, we don't
want to chown anything, no SELinux labeling, etc. There's no reason to write the
metadata to the `BARE_USER` xattrs.
The "unprivileged" context as used for `ex container` was a flag that was passed
down, but I think it's a lot easier if things just infer this state by looking
at whether the target repo is `bare-user-only`. Now we just have an
`rpmostree_context_new_tree()` API that handles both container and compose.
Closes: #1026
Approved by: jlebon
The previous change to log to the journal broke running `ex container` as
non-root with scripts. Yes, I really desperately need to set up real tests for
that, like we have for composes. And obviously checking `getuid() == 0` is a bit
crude but doing better would require plumbing through quite a bit of of
knowledge from the toplevel down into script execution.
Closes: #1021
Approved by: jlebon
Was just reading the initramfs bits while talking to rubao about kargs work, and
noticed we had a lot of "old style" declarations.
Closes: #1020
Approved by: jlebon
This is a quick band-aid to improve Ctrl-C handling in this area:
https://github.com/projectatomic/rpm-ostree/issues/897
At least this way we'll exit earlier rather than waiting until all the repos are
downloaded (and imported into libsolv, which is itself slow).
Closes: #1018
Approved by: jlebon
Sometimes it's useful to have access to the additional files when running
the post script, so this re-orders the compose process to copy the
additional files in before the post script runs
Signed-off-by: Jonathan Dieter <jdieter@lesbg.com>
Closes: #997
Approved by: jlebon
Even though it's really `/usr/etc`. This is for greater consistency with
`postprocess-script` where it appears as `/etc`.
Closes: #997
Approved by: jlebon
Lots of confusion in the codebase about this. The basic problem is that in
*most* cases, our code doesn't care; it's conceptually operating on `/usr/etc`,
which we could maintain as `/etc` and just rename it back at the very end.
The exceptions though are the `/etc/passwd` handling and livefs. And of course
libostree needs to handle `/usr/etc` vs `/etc` for config merging.
I considered trying to keep things the other way, but while I think we have some
ugly added here in this patch for things where we need to maintain an external
view (`remove-files` and `remove-from-packages`, and boy am I glad we had tests
for those), this ends up being mostly more consistent elsewhere.
One thing that might help is to maintain a fd for it; but that'd be an even more
invasive change.
This also ends up rolling in some unified core prep from
https://github.com/projectatomic/rpm-ostree/pull/940 in the form of
`rename_if_exists()` - basically for some minimal rootfs we may not have
`/boot`, or for that matter potentially even `etc`.
Prep for https://github.com/projectatomic/rpm-ostree/pull/997Closes: #997
Approved by: jlebon
I was reading some of the core code recently, and in general I think we need
more comments. I made a pass from the top of the file and got halfway down.
Closes: #1017
Approved by: jlebon
Rather than just letting the scriptlets inherit the daemon's
stdout/stderr, redirect their outputs so that we can set a customized
identifier to make it easier to distinguish from the daemon output.
Also print out the `journalctl` command needed so that users can
investigate the output themselves.
Closes: #998
Approved by: cgwalters
I suspect a common pattern with local replacement overrides is to
simultaneously replace a group of packages that depend on each other in
one shot, as is the case with docker, docker-common, and
docker-rhel-push-plugin currently in Fedora Atomic Host. In such cases,
we can print a cleaner diff in the status to make it easier to grok.
Before:
ReplacedBasePackages: strace 4.18-1.fc26 -> 4.19-1.fc26, docker-common 2:1.13.1-22.gitb5e3294.fc26 -> 2:1.13.1-21.git27e468e.fc26, docker 2:1.13.1-22.gitb5e3294.fc26 -> 2:1.13.1-21.git27e468e.fc26, docker-rhel-push-plugin 2:1.13.1-22.gitb5e3294.fc26 -> 2:1.13.1-21.git27e468e.fc26
After:
ReplacedBasePackages: docker-common docker docker-rhel-push-plugin 2:1.13.1-22.gitb5e3294.fc26 -> 2:1.13.1-21.git27e468e.fc26, strace 4.18-1.fc26 -> 4.19-1.fc26
Closes: #1004
Approved by: cgwalters
The comment here was wrong; we don't rely on `O_APPEND` here for package
layering since we convert on import. I noticed this while I was doing
a grep for `O_APPEND` in the codebase as part of unified core work.
Fix this by converting to `O_TMPFILE`+`GLNX_LINK_TMPFILE_NOREPLACE`.
Prep for unified core.
Closes: #1009
Approved by: jlebon
Prep for unified core work. This was failing due to a bug in libglnx
<https://github.com/GNOME/libglnx/pull/80> but I think this change is also
correct. There's no good reason for us to copy xattrs like the SELinux label
here - rather we want the labels to be reset during commit.
I did a tree-wide grep for other users and the only other case that is odd is
the treecompose `add-files`; I'd say we should change this but out of (a likely
excess of) conservatism I just left a "FIXME" for now.
Closes: #1008
Approved by: jlebon
Noticed this pass by while doing a `fedpkg local`. It doesn't actually
matter much since there are no exits possible between the point it's
declared and when it's actually assigned, which I will assume is why our
clang suite didn't pick it up, maybe? Not gonna look into that right
now.
Closes: #1010
Approved by: cgwalters
We were directly bind mounting the checked out `/usr/etc` onto `/etc`
which was exposing us to corruption from scriptlets. Since we already
have an rofiles-fuse mount for `/usr`, let's just re-use its `etc/`
subdir and bind mount that instead.
Closes: #1003
Approved by: cgwalters
We don't need those in the tree, so let's nuke them. This also fixes
subtle compatibility issues between hardlinks and lock files (see #999).
Closes: #1002
Approved by: cgwalters
Now that we support overrides, we need the sorter to make sure it goes
through the override packages in case they need to be relabeled.
Closes: #961Closes: #993
Approved by: cgwalters
Even if we're already initialized, we still need to create the subdir
requested. This would cause relabeling to fail because we try to create
the "relabel" subdir, which never actually gets created.
Closes: #993
Approved by: cgwalters
Amazingly we had 3 variations of this bug in the different
layers above using the `RpmOstreeContext` API (in compose, container,
and layering).
We fixed the first two already, this fixes the last one. We
get a warning if librpm holds a ref to a deleted rpmdb.
Closes: https://github.com/projectatomic/rpm-ostree/issues/987Closes: #994
Approved by: jlebon
This ensures we close the rpm database before we start moving
things around during postprocessing, and avoids a warning about
the rpmdb going missing being printed at the end of a compose.
Related: https://github.com/projectatomic/rpm-ostree/issues/987Closes: #989
Approved by: jlebon
This PR uses https://github.com/ostreedev/ostree/pull/1156
to ensure that when installing packages containing files with
exact same content, the files will be merged.
When installing packages containing conflicting files,
the error will still be detected at the ostree side.
The checkout overwrite option at rpm-ostree side is also
modified to accomodate the changes made in ostree side.
A test is added for regression
Commands like `upgrade` and `deploy` need to know if a new deployment
was actually laid down so that it may print a pkg diff if so. This is
implemented by listening for changes to the DefaultDeployment D-Bus
property. D-Bus emits a signal when the deployment variant changes
value.
However, in #595, with the introduction of `pending-*` related keys, the
deployment variant no longer represents data solely tied to that
specific deployment. In this case, because `deploy` operations currently
set the ref to the resolved checksum, it can happen that deploying the
same base commit when the current refspec *isn't* pointing to that base
commit will result in the `pending-*` keys dropping out and a default
deployment change notification going out.
In this patch, we strengthen how we determine whether a new deployment
was laid down by actually looking at the deployment id, rather than just
assuming that a change to the property implies a new deployment.
Closes: #981Closes: #984
Approved by: cgwalters
Also something I noticed while working on #981. When sitting on a livefs
commit, once a user does `rpm-ostree cleanup --pending --rollback`, it's
impossible to redeploy the same booted commit. Let's allow users to do
this.
Closes: #984
Approved by: cgwalters
Currently, when setting the `override-commit` key in the origin, the
upgrader pulls that commit checksum directly and then updates the
refspec to point to it. This behaviour was inherited from its ostree
version; at the time it was implemented, the pull code didn't support
passing a specific commit for a given refspec. However, we now have
the override-commit-ids option, which will make libostree update the ref
for us.
We change the code here to make use of it and simplify the function.
This also fixes the corner case of local branches: we shouldn't change
the ref if we're on a local branch. This is actually what drove me to
this patch as I was debugging #981.
(Aside: I'm still not convinced updating the refspec is always the
correct thing to do even in the remote case, though it's a bit messy to
disentangle).
Closes: #984
Approved by: cgwalters
Prep for changing `boot_location: new` to use `/usr/lib/ostree-boot`
and `/usr/lib/modules`. Rework our kernel postprocessing
so that we unify the `boot_location` handling with initramfs generation.
Instead of doing the initramfs first in postprocessing, we do it nearly last,
after e.g. `etc` is renamed to `usr/etc`. This has some consequences, such as
the fact that `run_bwrap_mutably()` is now called in both situations. In
general, our handling of `etc` is inconsistent, although understandably so.
As part of this, I finally got around to implementing the bit from
https://github.com/systemd/systemd/pull/4174 however suboptimal it is; need the
unified core so we can cleanly ignore the posttrans like we do others. We
intentionally keep the file around in the generated tree so that installing a
kernel RPM per client doesn't try to do any of this either.
This all gets folded together so that the logic for handling the bootloader gets
simpler - in the Fedora case, we now know to find kernels in `/usr/lib/modules`
and can ignore `/boot`.
Closes: #959
Approved by: jlebon
We have 3 locations to find kernels now; I can't think of
a reason to support placing kernels *only* in `/boot`. The
original commit
15ecaacd36
doesn't give a reason, and I certainly can't think of one now.
This makes `legacy` be an alias for `both`, which should be fully compatible.
Prep for further refactoring towards changing `new` to mean both
`/usr/lib/ostree-boot` *and* `/usr/lib/modules`.
Closes: #959
Approved by: jlebon
Make `upgrade` use the same trick as `deploy` to determine if a new
deployment was laid down. Apart from those two, all other operations
that can lay down a new deployment always do so in the happy path.
Prep for further work.
Closes: #983
Approved by: cgwalters
Today in Fedora the `glibc-all-langpacks.posttrans` is implemented
in lua, for no good reason. See:
https://bugzilla.redhat.com/show_bug.cgi?id=1367585
Since that's stalled out, let's add support for overrides. This
is obviously a much bigger step with more long term maintenance
implications over our current "ignore scripts" list. But we can't
block either.
This is needed for unified core work:
https://github.com/projectatomic/rpm-ostree/issues/729
(We also override `fedora-release-atomichost` but I'll likely
submit a patch for that upstream)
Closes: #980
Approved by: jlebon
I was linking to this code from elsewhere and noticed that
for our hardlink breaks we were not using fd-relative even
though we can. Down the line if we fork librpm into a separate
process and do e.g. `--dbpath=.` it'll do it too.
(Side note, I verified that commenting out the hardlink breaking
here was caught by the `ostree fsck` I added to the test suite)
Closes: #979
Approved by: jlebon
Started on porting to the new `glnx_fstatat_allow_noent()`. The
usage varies a lot and it felt easy to screw up, so I'm just
starting by doing a few of them.
Update submodule: libglnx
Closes: #978
Approved by: jlebon
I'd like to cut down on the number of tempdirs we make; when
debugging with gdb I often use e.g. `r` to restart forcibly, and
that leaves the dirs around.
Previously we'd make a tmpdir per relabeled package; this hoists it into a
well-known "relabel/" subdir of the per-core tmpdir we make. (A next step here
is to dedup the core tmpdir with the tmpdir that the upgrader makes)
Further while I'm here, let's use a single OSTree transaction for all of the
relabeling; the rationale here is that it's better to do fewer `syncfs()`
operations.
Closes: #970
Approved by: jlebon
This fixes `rpm-ostree reload` as root, and supports configuring
it to be enabled for other users as well. This was overlooked
in the polkit work originally.
Closes: https://github.com/projectatomic/rpm-ostree/issues/976Closes: #977
Approved by: jlebon
At one point `rpm-ostree install libvirt` dragged in libguestfs which in turn
brought in `syslinux-extlinux-nonlinux` which has files in `/boot/extlinux`,
which we rejected. (That dependency chain appears to have been fixed currently)
For the general case, this is just a partial fix in that we haven't nailed down
the semantics of how updates for `/boot` work. But in this particular case,
we'll just break libguestfs' `extlinux` verb, which I'm OK with.
Another case is `fwupdate-efi` - we require manual intervention to copy the
data into `/boot` after installing the package.
This is also preparation for [unified core](https://github.com/projectatomic/rpm-ostree/issues/729)
in that we now ensure imported kernels don't end up in `/boot` unless
explicitly configured.
Closes: https://github.com/projectatomic/rpm-ostree/issues/853Closes: #969
Approved by: jlebon
I noticed our C-style comments were actually showing up in the generated hash
table. Consistently use gperf's `#`, and also move all comments to the line
above the value for more readability.
Closes: #967
Approved by: jlebon
While working on unified core and the Fedora Atomic Host content set, I hit a
dependency between `docker.posttrans` which tries to read `/etc/os-release`, and
`fedora-release-atomichost.post` which creates that symlink.
It seems best practice to me to run `%post`s strictly before
`%posttrans`; we're not likely to do parallelization anytime
soon anyways.
While here I cleaned things up by having an enum for the script kind,
rather than multiple functions, otherwise we would have had another
wrapper in core.c.
Closes: #963
Approved by: jlebon
I initially thought I wouldn't need this but it turns out it makes things nicer
unified core. In the legacy path we'll need to write
`etc/passwd`, but now we need `usr/etc/passwd` in the unified path, and
we can start to match up with the other code paths which are fd-relative.
Closes: #950
Approved by: jlebon
The core use case always enabled this; we exposed it on the command line, but
honestly I think it's a lot simpler if our imports always enable this. If we
need to *undo* it (for example, generating a "traditional feeling" OCI/Docker
image a la <https://github.com/projectatomic/rpm-ostree/pull/367>), we can
easily do that as postprocessing afterwards.
Closes: #951
Approved by: jlebon
We should be using `g_propagate_error()` to ensure we handle the case where the
caller specifies `NULL` for the error.
Closes: #951
Approved by: jlebon
Just noticed this while trying out the unified core work. We were
passing arguments in the wrong order (or alternatively, receiving them
in the wrong order).
Closes: #952
Approved by: cgwalters
This is a trap when doing style porting; we added a rule to ostree's `maint.mk`
to catch it but not here. Not sure I want to add `maint.mk` here yet.
Note that in newer libglnx `glnx_opendirat()` does error prefixing, so
drop the prefixing there.
Closes: #943
Approved by: jlebon
It was partially removed in 325ee354e9
from PR #873 - this removes all of the rest of the layers in
the daemon and core for this.
Note that while working on this I found it astoundingly confusing that
`rpmts_add_install()` and `add_install()` had separate orderings for `gboolean
noscripts, is_install`. That's fixed by having the latter lose the `noscripts`
flag entirely, but while we're here change `rpmts_add_install()` to use flags
for clarity.
Closes: #938
Approved by: jlebon
Missed this in the tmpdir porting since it used `g_mkdtemp_full()`.
We also now have autocleanups for `OstreeRepoCommitModifier`.
Closes: #939
Approved by: jlebon
I was about to copy this bit from the daemon for subprocess work, so let's
centralize it. Further, we should invoke `setenv()` as early as possible; see
<https://sourceware.org/bugzilla/show_bug.cgi?id=15607#c2>.
Closes: #934
Approved by: jlebon
See <https://github.com/projectatomic/bubblewrap/pull/165>; really every
bwrap use case I can think of should specify this (including ours), it's
just not the default out of conservatism.
This way if the daemon happens to e.g. SEGV it'll also cleanly `SIGKILL` any
outstanding scripts.
Was just looking at our bwrap usage for multiprocess work.
Closes: #936
Approved by: jlebon
Lots of cleanups, and this also allows us to fully port to new style in several
places.
We can drop our own `rpmostree_mkdtemp()`.
Update submodule: libglnx
Closes: #928
Approved by: jlebon
Since we have a copy of this libostree code, pick up the new
changes from <https://github.com/ostreedev/ostree/pull/1055>.
Note the added test doesn't really test our logic since
we're only doing local pulls, but at least we have something.
Closes: #932
Approved by: jlebon
There's no reason to touch things in the target root *before*
we've even downloaded the repodata. Further, doing so has
the potential trigger the "existing root" logic in libdnf if
it detects `/usr`. This doesn't matter now since we're writing
to `/etc/`, but it will for later work, so let's put these bits
in the right place.
Prep for <https://github.com/projectatomic/rpm-ostree/pull/927>
Closes: #931
Approved by: jlebon
systemd has to go to a bit more effort to integrate `syslog()`, let's
use the journal directly for consistency. Just happened to be
reading the code.
Closes: #921
Approved by: jlebon
Make use of the new journal-msg signal that OstreeSysroot emits. Since
we now also only rely on the output API to print messages, this allows
us to drop the outrageous hacks we had to do to watch our own stdout,
part of which Coverity didn't like.
Coverity CID: 163694
Closes: #919
Approved by: cgwalters
There were places where we were relying on the stdout self read hack to
relay messages to clients. Fix those by explicitly using the output API
in all spots where the daemon may pass. Really, we should exclusively be
using the output API in the daemon and libpriv trees, though many other
instances exist for now.
Closes: #919
Approved by: cgwalters
This closes a longstanding bug - since package layering first
landed, we only checked for newer RPMs if the base tree changed.
In some scenarios like RHELAH, this doesn't matter much by default
since they move at the same cadence. Except if you use EPEL for example.
In Fedora, today the FAH releases are async of the rpm-md repos, and
there's also COPR which can update more than once a day even.
We should check for both update sources. Luckily we'd already introduced logic
for this in the treecompose case (checksumming the depsolved package sack). We
just need to start using it for client side assembly too.
Closes: https://github.com/projectatomic/rpm-ostree/issues/391Closes: #911
Approved by: jlebon
`g_mkdtemp_full` returns a pointer to the template var itself or NULl on
error. We were treating it like an integer.
Coverity CID: 163698
Closes: #918
Approved by: cgwalters
I hesitated on this one. Really, we *could* just explicitly cast the
result to mark it as used, though OTOH, we really should at least drop
something in the journal. It's not fatal, but it shouldn't happen so we
should know about it.
Coverity CID: 163717
Closes: #918
Approved by: cgwalters
1. We weren't initializing all fds to -1, but 0 is a valid fd number.
2. We were checking for nonzero rather than explicitly for -1.
Coverity CID: 163691
Coverity CID: 163700
Coverity CID: 163714
Closes: #918
Approved by: cgwalters
Before, when using rpm-ostree -h command, the command
description was shown together with the command itself
Now, we separate the command description out, providing
user a better view of the description.
Closes: #916
Approved by: jlebon
This is no longer necessary since we switched to tracking livefs state in the
origin, and we don't support changing content outside of `/usr` right now
via livefs.
Just noticed this while looking at the livefs code.
Closes: #912
Approved by: jlebon
Move basically everything except early option
processing into a helper function.
Also update a few other misc bits above that to new style.
Closes: #904
Approved by: jlebon
Down the line when we have a unified core, this will be totally unnecessary.
Even without that, it's not worth the maintenance pain for us. Anyone who wants
this can easily on their own mount a tmpfs and point us at it with --workdir.
Prep for further code cleanup.
Closes: #904
Approved by: jlebon
