If the /var/log/journal directory is created with rigths 700, the application
of an ACL rules without any primary group right sets it to 0. A chmod 755 on
this file will then only set the ACL mask and let the ACL primary group right
to 0. The directory is then unreadable for the primary group.
This patch explicitly sets the primary group to avoid this problem.
Fixes#5264.
Let's automatically destory per-unit private temporary directories, as
they are created by PrivateTmp=yes on each boot, if we notice them to be
around, in case they are left-overs from the last boot.
Fixes: #4401
In order to improve compatibility with local clients that speak DNS directly
(and do not use NSS or our bus API) listen locally on 127.0.0.53:53 and process
any queries made that way.
Note that resolved does not implement a full DNS server on this port, but
simply enough to allow normal, local clients to resolve RRs through resolved.
Specifically it does not implement queries without the RD bit set (these are
requests where recursive lookups are explicitly disabled), and neither queries
with DNSSEC DO set in combination with DNSSEC CD (i.e. DNSSEC lookups with
validation turned off). It also refuses zone transfers and obsolete RR types.
All lookups done this way will be rejected with a clean error code, so that the
client side can repeat the query with a reduced feature set.
The code will set the DNSSEC AD flag however, depending on whether the data
resolved has been validated (or comes from a local, trusted source).
Lookups made via this mechanisms are propagated to LLMNR and mDNS as necessary,
but this is only partially useful as DNS packets cannot carry IP scope data
(i.e. the ifindex), and hence link-local addresses returned cannot be used
properly (and given that LLMNR/mDNS are mostly about link-local communication
this is quite a limitation). Also, given that DNS tends to use IDNA for
non-ASCII names, while LLMNR/mDNS uses UTF-8 lookups cannot be mapped 1:1.
In general this should improve compatibility with clients bypassing NSS but
it is highly recommended for clients to instead use NSS or our native bus API.
This patch also beefs up the DnsStream logic, as it reuses the code for local
TCP listening. DnsStream now provides proper reference counting for its
objects.
In order to avoid feedback loops resolved will no silently ignore 127.0.0.53
specified as DNS server when reading configuration.
resolved listens on 127.0.0.53:53 instead of 127.0.0.1:53 in order to leave
the latter free for local, external DNS servers or forwarders.
This also changes the "etc.conf" tmpfiles snippet to create a symlink from
/etc/resolv.conf to /usr/lib/systemd/resolv.conf by default, thus making this
stub the default mode of operation if /etc is not populated.
When ACL support is enabled, systemd-tmpfiles-setup service sets the following
ACL entries to the volatile system journal:
$ getfacl /run/log/journal/*/system.journal
getfacl: Removing leading '/' from absolute path names
# file: run/log/journal/xxx/system.journal
# owner: root
# group: systemd-journal
user::rwx
group::r--
group🛞r-x
group:adm:r-x
mask::r-x
other::---
This patch makes sure that the exec bit is not set anymore for the volatile
system journals.
Hardly any software uses that any more, and better locking mechanisms like
flock() have been available for many years.
Also drop the corresponding "lock" group from sysusers.d/basic.conf.in, as
nothing else is using this.
This way, directories created later for containers or for
journald-remote, will be readable by adm & wheel groups by default,
similarly to /var/log/journal/%m itself.
https://github.com/systemd/systemd/issues/1971
/etc/mtab should be labeled as "_", even though systemd has its own
smack label using '--with-smack-run-label' configuration. This is mainly
because all processes could read that file and the origin of this file
(i.e. /proc/mounts) is labeled as "_". This labels /etc/mtab as "_" when
'--with-smack-run-label' is enabled.
Do so only in /run. We shouldn't alter ACLs for existing files in /var,
but only for new files. If the admin made changes to the ACLs they
shouls stay in place.
We should still do recursive ACL changes for files in /run, since those
are not persistent, and will hence lack ACLs on every boot.
Also, /var/log/journal might be quit large, /run/log/journal is usually
not, hence we should avoid the recursive descending on /var, but not on
/run.
Fixes#534
Remove old temporary snapshots, but only at boot. Ideally we'd have
"self-destroying" btrfs snapshots that go away if the last last
reference to it does. To mimic a scheme like this at least remove the
old snapshots on fresh boots, where we know they cannot be referenced
anymore. Note that we actually remove all temporary files in
/var/lib/machines/ at boot, which should be safe since the directory has
defined semantics. In the root directory (where systemd-nspawn
--ephemeral places snapshots) we are more strict, to avoid removing
unrelated temporary files.
This also splits out nspawn/container related tmpfiles bits into a new
tmpfiles snippet to systemd-nspawn.conf
We will create the symlink on boot as a fallback to provide name
resolution. But if the symlink was removed afterwards, it most likely
should not be recreated. Creating it only on boot also solves the
issue where it would be created prematurely during installation,
before the system was actually booted.
https://bugzilla.redhat.com/show_bug.cgi?id=1197204
Add the +C file attribute (NOCOW) to the journal directories, so that
the flag is inherited automatically for new journal files created in
them. The journal write pattern is problematic on btrfs file systems as
it results in badly fragmented files when copy-on-write (COW) is used:
the performances decreases substantially over time.
To avoid this issue, this tmpfile.d snippet sets the NOCOW attribute to
the journal files directories, so newly created journal files inherit
the NCOOW attribute that disables copy-on-write.
Be aware that the NOCOW file attribute also disables btrfs checksumming
for these files, and thus prevents btrfs from rebuilding corrupted files
on a RAID filesystem.
In a single disk filesystems (or filesystems without redundancy) it is
safe to use the NOCOW flags without drawbacks, since the journal files
contain their own checksumming.
This patch removes unnecessary blank line in
/usr/lib/tmpfiles.d/etc.conf when configured with "--disable-resolved".
(i.e. ENABLE_RESOLVED is not defined)
Given that this is also the place to store raw disk images which are
very much bootable with qemu/kvm it sounds like a misnomer to call the
directory "container". Hence, let's change this sooner rather than
later, and use the generic name, in particular since we otherwise try to
use the generic "machine" preferably over the more specific "container"
or "vm".
Choose which system users defined in sysusers.d/systemd.conf and files
or directories in tmpfiles.d/systemd.conf, should be provided depending
on comile-time configuration.
Create /var/lib/containers so that it exists with an appropriate mode. We want
0700 by default so that users on the host aren't able to call suid root
binaries in the container. This becomes a security issue if a user can enter a
container as root, create a suid root binary, and call that from the host.
(This assumes that containers are caged by mandatory access control or are
started as user).
Now that logind will clean up all IPC resources of a user we should
really consider $XDG_RUNTIME_DIR as just another kind of IPC with the
same life-cycle logic as the other IPC resources. This should be safe
now to do since every user gets his own $XDG_RUNTIME_DIR tmpfs instance
with a fixed size limit, so that flooding of it will more effectively be
averted.
Management of /var/cache/man should move to the distribution package
owning the directory (for example, man-db). As man pages are a
non-essential part of the system and unnecessary for minimal setups,
there's no point in having systemd ship these lines.
Distribution packages should make sure the appropriate package for their
distribution adopts this line. Ideally, the line is adopted by the
upstream package.
For Fedora I have filed this bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1110274
"m" so far has been a non-globbing version of "z". Since this makes it
quite redundant, let's get rid of it. Remove "m" from the man pages,
beef up "z" docs instead, and make "m" nothing more than a compatibility
alias for "z".
Configuration will be in
root:root /run/systemd/network
and state will be in
systemd-network:systemd-network /run/systemd/netif
This matches what we do for logind's seat/session state.
Various operations done by systemd-tmpfiles may only be safely done at
boot (e.g. removal of X lockfiles in /tmp, creation of /run/nologin).
Other operations may be done at any point in time (e.g. setting the
ownership on /{run,var}/log/journal). This distinction is largely
orthogonal to the type of operation.
A new switch --unsafe is added, and operations which should only be
executed during bootup are marked with an exclamation mark in the
configuration files. systemd-tmpfiles.service is modified to use this
switch, and guards are added so it is hard to re-start it by mistake.
If we install a new version of systemd, we actually want to enforce
some changes to tmpfiles configuration immediately. This should now be
possible to do safely, so distribution packages can be modified to
execute the "safe" subset at package installation time.
/run/nologin creation is split out into a separate service, to make it
easy to override.
https://bugzilla.redhat.com/show_bug.cgi?id=1043212https://bugzilla.redhat.com/show_bug.cgi?id=1045849
This way it is easy to only exclude directories from the current boot
from automatic clean up in /var/tmp.
Also, pick a longer name for the directories so that are globs in
tmp.conf can be simpler yet equally accurate.
In order to avoid a deadlock between journald looking up the
"systemd-journal" group name, and nscd (or anyother NSS backing daemon)
logging something back to the journal avoid all NSS in journald the same
way as we avoid it from PID 1.
With this change we rely on the kernel file system logic to adjust the
group of created journal files via the SETGID bit on the journal
directory. To ensure that it is always set, even after the user created
it with a simply "mkdir" on the shell we fix it up via tmpfiles on boot.
Embedded folks don't need the machine registration stuff, hence it's
nice to make this optional. Also, I'd expect that machinectl will grow
additional commands quickly, for example to join existing containers and
suchlike, hence it's better keeping that separate from loginctl.
Currently, PrivateTmp=yes means that the service cannot see the /tmp
shared by rest of the system and is isolated from other services using
PrivateTmp, but users can access and modify /tmp as seen by the
service.
Move the private /tmp and /var/tmp directories into a 0077-mode
directory. This way unpriviledged users on the system cannot see (or
modify) /tmp as seen by the service.
We finally got the OK from all contributors with non-trivial commits to
relicense systemd from GPL2+ to LGPL2.1+.
Some udev bits continue to be GPL2+ for now, but we are looking into
relicensing them too, to allow free copy/paste of all code within
systemd.
The bits that used to be MIT continue to be MIT.
The big benefit of the relicensing is that closed source code may now
link against libsystemd-login.so and friends.
This extends the shutdownd interface to expose schedule shutdown
information in /run/systemd/shutdown/schedule.
This also cleans up the shutdownd protocol and documents it in a header
file sd-shutdown.h.
This is supposed to be used by client code that wants to control and
monitor scheduled shutdown.
Many people prefer to avoid clearing /tmp and /var/tmp, and
distributions often have explicit settings for how often to clear them
if at all. Overriding those with systemd currently requires overriding
all of /usr/lib/tmpfiles.d/systemd.conf via
/etc/tmpfiles.d/systemd.conf, copying across all the other entries, and
updating that override when systemd.conf changes.
Move the /tmp and /var/tmp entries from systemd.conf to a separate
tmp.conf, making them easier to override without affecting the rest of
systemd.conf.
Instead of the /dev/.run trick we have currently implemented, we decided
to move the early-boot runtime dir to /run.
An existing /var/run directory is bind-mounted to /run. If /var/run is
already a symlink, no action is taken.
An existing /var/lock directory is bind-mounted to /run/lock.
If /var/lock is already a symlink, no action is taken.
To implement the directory vs. symlink logic, we have a:
ConditionPathIsDirectory=
now, which is used in the mount units.
Skipped mount unit in case of symlink:
$ systemctl status var-run.mount
var-run.mount - Runtime Directory
Loaded: loaded (/lib/systemd/system/var-run.mount)
Active: inactive (dead)
start condition failed at Fri, 25 Mar 2011 04:51:41 +0100; 6min ago
Where: /var/run
What: /run
CGroup: name=systemd:/system/var-run.mount
The systemd rpm needs to make sure to add something like:
%pre
mkdir -p -m0755 /run >/dev/null 2>&1 || :
or it needs to be added to filesystem.rpm.
Udev -git already uses /run if that exists, and is writable at bootup.
Otherwise it falls back to the current /dev/.udev.
Dracut and plymouth need to be adopted to switch from /dev/.run to run
too.
Cheers,
Kay
lewo