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.
'systemctl cat' now works for templates too.
'systemctl edit' does not refuse to edit units that have changed on
disk. That restriction didn't seem useful, actually editing units that
have changed on disk before they are started is very reasonable.
'edit' with instances and templates works again:
Now:
$ build/systemctl edit getty@
Failed to copy /etc/systemd/system/getty@.service.d/override.conf to /etc/systemd/system/getty@.service.d/.override.confdff6290408c86369: Permission denied
$ build/systemctl edit getty@tty3
Failed to create directories for /etc/systemd/system/getty@tty3.service.d/override.conf: Permission denied
$ build/systemctl edit --full getty@tty3
Failed to copy /usr/lib/systemd/system/getty@.service to /etc/systemd/system/.getty@tty3.serviced3d175087e7e439b: Permission denied
Failed to create temporary file for /etc/systemd/system/getty@tty3.service: Permission denied
$ build/systemctl edit --full getty@
Failed to copy /usr/lib/systemd/system/getty@.service to /etc/systemd/system/.getty@.servicea3caad491c0f2f3d: Permission denied
Failed to create temporary file for /etc/systemd/system/getty@.service: Permission denied
Containers do not really support .device, .automount or .swap units;
Systems compiled without support for swap do not support .swap units;
Systems without kdbus do not support .busname units.
With this change attempts to start a unsupported unit types will result
in an immediate "unsupported" job result, which is a lot more
descriptive then before. Also, attempts to start device units in
containers will now immediately fail instead of causing jobs to be
enqueued that never go away.
Previously, if we provided getty@.service to systemctl edit it would
have failed when using the bus because it is an invalid unit name.
But it would have succeeded when searching in the filesystem.
Now, we check if we have a template, if we do we search in the
filesystem, if we don't have a templae and we can use the bus, we do.
Furthermore, if we provided getty@tty1.service it would not have worked
when searching the filesystem, but it would have worked with the bus.
So now, when using the filesystem we use the template name and not the
unit name, and the same when logging errors.
(Also did a refactoring to avoid a long function)
try_context() is such a hot path that the hashmap lookup is expensive.
The number of contexts is small - it is the number of object types.
Using a hashmap is overkill. A plain array will do.
Before:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 0m9.445s
user 0m9.228s
sys 0m0.213s
After:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 0m5.438s
user 0m5.266s
sys 0m0.170s
Note that numbers 0 and -1 are both replaced with OBJECT_UNUSED,
because they are treated the same everywhere (e.g. type_to_context()
translates them both to 0).
If type==0 and a non-NULL object were given as arguments to
journal_file_hmac_put_object(), its object type check would fail and it
would return -EBADMSG.
All existing callers use either a positive type or -1. Still, for
behavior consistency with journal_file_move_to_object() let's allow
type 0 to pass.
The only user is sd_journal_enumerate_unique() and, as explained in
the previous commit (fed67c38e3 "journal: map objects to context set by
caller, not by actual object type"), the use of them there is now
superfluous. Let's remove them.
This reverts major parts of commits:
ae97089d49 journal: fix access to munmapped memory in
sd_journal_enumerate_unique
06cc69d44c sd-journal: fix sd_journal_enumerate_unique skipping values
Tested with an "--enable-debug" build and "journalctl --list-boots".
It gives the expected number of results. Additionally, if I then revert
the previous commit ("journal: map objects to context set by caller, not
to actual object type"), it crashes with SIGSEGV, as expected.
When the caller of journal_file_move_to_object() specifies type==0,
the object header is at first mapped in context 0. Then after the header
is checked, the whole object is mapped in a context determined by
the actual object type (which is not even range-checked using
type_to_context()). This looks wrong. It should map in the
caller-specified context.
An old comment in sd_journal_enumerate_unique() supports this view:
/* We do not use the type context here, but 0 instead,
* so that we can look at this data object at the same
* time as one on another file */
Clearly the expectation was that the data object will remain mapped
in context 0 without being pushed away by mapping other objects in
context OBJECT_DATA.
I suspect that this was the real bug that got fixed by ae97089d49
"journal: fix access to munmapped memory in sd_journal_enumerate_unique".
In other words, journal_file_object_keep/release are superfluous after
applying this patch.
This is useful for exposing unsafe access to mmapped objects after
the context that they were mapped in was already moved.
For example:
journal_file_move_to_object(f1, OBJECT_DATA, p1, &o1);
journal_file_move_to_object(f2, OBJECT_DATA, p2, &o2);
t = o1->object.type; /* this usually works, but is unsafe */
There will be more debugging options later.
--enable-debug will enable them all.
--enable-debug=hashmap will enable only hashmap debugging.
Also rename the C #define to ENABLE_DEBUG_* pattern.
Also, when booting up an ephemeral container of / use the system
hostname as default machine name.
This way specifiyng -M is unnecessary when booting up an ephemeral
container, while allowing any number of ephemeral containers to run from
the same tree.
This works now:
# systemd-nspawn -xb -D / -M foobar
Which boots up an ephemeral container, based on the host's root file
system. Or in other words: you can now run the very same host OS you
booted your system with also in a container, on top of it, without
having it interfere. Great for testing whether the init system you are
hacking on still boots without reboot the system!
This adds --template= to duplicate an OS tree as btrfs snpashot and run
it
This also adds --ephemeral or -x to create a snapshot of an OS tree and
boot that, removing it after exit.
Also, rename filename_is_safe() to filename_is_valid(), since it
actually does a full validation for what the kernel will accept as file
name, it's not just a heuristic.
Imagine the following use of hwdb:
if (condition_A)
SD_HWDB_FOREACH_PROPERTY(hwdb, modalias, key, value)
operation_A(key, value);
else
log_error("...");
This should work just fine, but but definitely does not what you would
expect. Due to how SD_HWDB_FOREACH_PROPERTY is defined, the dangling
'else' is linked to the hidden 'if' statement in the macro instead of the
outer 'if (condition_A)'. This is unexpected and really annoying to debug.
Fix this by never leaving un-finished if-statements in
SD_HWDB_FOREACH_PROPERTY(). We simply inverse the if() statement and
explicitly add an 'else'-branch. This way, the statement is closed and all
ambuguities are resolved.
This is libudev-hwdb, but decoupled from libudev and in the libsystemd style.
The core code is unchanged, apart from the following minor changes:
- hwdb.bin located in /**/systemd/hwdb/ take preference over the ones located
in /**/udev/
- properties are stored internally in an OrderedHashmap, rather than a
linked list.
- a new API call allows individual properties to be queried directly, rather
than iterating over them all
- the iteration over properties have been moved inside the library, rather than
exposing a list directly
- the unused 'flags' parameter was dropped
Sync up with recent kdbus changed:
* several ioctls gained .size and .items members (but still unused)
* CMD_SEND gained its own ioctl structure
* several members of kdbus_msg were dropped as they were only used during
SEND, not during RECV etc.
* CMD_RECV and CMD_SEND now share a kdbus_reply member which contains the
offset and size of the returned message.
We must restore part->mmap_begin when poping memfds from the memfd-cache.
We rely on the memfds to be unsealed, so we can be sure that we own the
whole FD. Therefore, simply set part->mmap_begin to the same as
part->data.
This fixes test-bus-kernel-benchmark.
Let's stick to generic sections that describe the general technology,
instead of specific per-object sections, unless we really have a reason
to do that otherwise.
When ICMPv6 Other information is received, enable Information request
in DHCPv6. If the DHCPv6 client already exists, only update the client
if there is a transition from Other to Managed state.
Start the DHCPv6 test case by sending an Information Request, verifying
the reply and continuing at once with the normal address acquisition
procedure. Reuse the DHCPv6 Solicit Reply so that the client code is
verified to ignore any erroneously added IPv6 address information.
Implement Information Request message according to RFC 3315, section
18.1.5. with the excepion that the first message is not delayed by a
random amount. Instead systemd-networkd is supposed to take care of
desynchronizing between other clients.
Initialize the DHCPv6 client structure in sd_dhcp6_client_start()
as this allows toggling between information request and normal
DHCPv6 address aquisition modes.
Suppyling a NULL lease is not a condition that needs to be asserted,
returning -EINVAL is informative enough to the caller. This simplifies
calling code and doesn't falsely indicate that something erroneous was
attempted.
When all DHCPv6 options have been parsed, dhcp6_option_parse() returns
-ENOMSG. Explicitely set the return value to indicate success so that
later code does not need to take this special value into account.
This way, we should be in a slightly better situation if a container is
booted up with only a shell as PID 1. In that case
/run/systemd/container will not be populated, and a check for it hence
be ineffective.
Checking /proc/1/environ doesn't fully fix the problem though, as the
file is only accessible with privileges. This means if PID 1 is not
systemd, and if privileges have been dropped the container detection
will continue to fail.
The error was introduced in v215-343-g60731f32f1 'systemctl: do not
bother to mutate state on error', by causing strv_free to attempt to
free a static string. Simplify the whole thing by always keeping the
array in valid state.
loop_write() didn't follow the usual systemd rules and returned status
partially in errno and required extensive checks from callers. Some of
the callers dealt with this properly, but many did not, treating
partial writes as successful. Simplify things by conforming to usual rules.
Suppose that while iterating we have already looked into a journal file
and got a candidate for the next entry. And we are considering to look
into another journal file because it may contain an entry that is nearer
to the current location than the candidate.
We should skip the whole journal file if we can tell by looking at its
header that none of its entries can precede the candidate.
Before:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 0m20.518s
user 0m19.989s
sys 0m0.328s
After:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 0m9.445s
user 0m9.228s
sys 0m0.213s
Interleaving of entries from many journal files is expensive. But there
is room for optimization.
We can skip looking into journal files whose entries all lie before the
current iterating location. We can tell if that's the case from looking
at the journal file header. This saves a huge amount of work if one has
many of mostly not interleaved journal files.
On my workstation with 90 journal files in /var/log/journal/ID/
totalling 3.4 GB I get these results:
Before:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 5m54.258s
user 2m4.263s
sys 3m48.965s
After:
$ time ./journalctl --since=2014-06-01 --until=2014-07-01 > /dev/null
real 0m20.518s
user 0m19.989s
sys 0m0.328s
The high "sys" time in the original was caused by putting more stress on
the mmap-cache than it could handle. With the patch the working set
now consists of fewer mmap windows and mmap-cache is not thrashing.
If a unit is set property by "systemctl set-property", a new dropin
file is generated. But the unit's dropin_paths and dropin_mtime are
not updated. So the unit is shown as need daemon reload.
Update unit dropin_paths and dropin_mtime also when dropin file is
written.
When dbus client connects to systemd-bus-proxyd through
Unix domain socket proxy takes client's smack label and sets for itself.
It is done before and independent of dropping privileges.
The reason of such soluton is fact that tests of access rights
performed by lsm may take place inside kernel, not only
in userspace of recipient of message.
The bus-proxyd needs CAP_MAC_ADMIN to manipulate its label.
In case of systemd running in system mode, CAP_MAC_ADMIN
should be added to CapabilityBoundingSet in service file of bus-proxyd.
In case of systemd running in user mode ('systemd --user')
it can be achieved by addition
Capabilities=cap_mac_admin=i and SecureBits=keep-caps
to user@.service file
and setting cap_mac_admin+ei on bus-proxyd binary.
Support timer options --on-active=, --on-boot=, --on-startup=,
--on-unit-active=, --on-unit-inactive=, --on-calendar=. Each options
corresponding with OnActiveSec=, OnBootSec=, OnStartupSec=,
OnUnitActiveSec=, OnUnitInactiveSec=, OnCalendar= of timer
respectively. And OnCalendar= and WakeSystem= supported by
--timer-property= option like --property= of systemd-run.
And if --unit= option and timer options are specified the command can
be omitted. In this case, systemd-run assumes the target service is
already loaded. And just try to generate transient timer unit only.
kdbus-git gained two new features:
* memfd offsets: This allows to specify a 'start' offset in kdbus_memfd
so you can send partial memfd hunks instead of always
the full memfd
* KDBUS_HELLO_UNPRIVILEGED: If passed during HELLO, the client will be
treated as unprivileged.
In the case where no entries have been added to the journal after the specified
cursor, set need_seek before the main loop to prevent display of the entry at
said cursor.
If a cgroup fails to be destroyed (most likely because there are still
processes running as part of a service after the main pid exits), don't
free and remove the cgroup unit from the manager. This fixes a
regression introduced by the cgroup rework in v205 where systemd would
forget about processes still running after the unit becomes inactive.
(This can happen when the main pid exits and KillMode=process or none).
There is no need to require mount.usrflags. The original implementation
assumed that a btrfs subvolume would always be needed but that is not
applicable to systems that do not use btrfs for /usr.
Similar to using rootflags= for the default of mount.usrflags=, append
the classic 'ro' and 'rw' flags to the mount options.
With DIRECTION_UP (i.e. navigating backwards) in generic_array_bisect() when the
needle was found as the last item in the array, it wasn't actually processed as
match, resulting in entries being missed.
https://bugs.freedesktop.org/show_bug.cgi?id=86855
The ELF magic cannot work for consumers of our shard library, since they
are in a different module. Hence make all the ELF magic private, and
instead introduce a public function to register additional static
mapping table.
Implement a recent change in the kdbus pool logic:
PAYLOAD_VEC_OFF items are now referencing offsets relative to the
connection's pool, not to the item itself. Follow this change in
sd-bus.
This patch makes it possible to set extended attributes on files created
by tmpfiles. This can be especially used to set SMACK security labels on
volatile files and directories.
It is done by adding new line of type "t". Such line should contain
attributes in Argument field, using following format:
name=value
All other fields are ignored.
If value contains spaces, then it must be surrounded by quotation marks.
User can also put quotation mark in value by escaping it with backslash.
Example:
D /var/run/cups - - - -
t /var/run/cups - - - - security.SMACK64=printing
It's only exposed to userspace since
commit 685343fc3ba61a1f6eef361b786601123db16c28
Author: Tom Gundersen <teg@jklm.no>
AuthorDate: Mon Jul 14 16:37:22 2014 +0200
Commit: David S. Miller <davem@davemloft.net>
CommitDate: Tue Jul 15 16:12:01 2014 -0700
to the kernel.
This has been requested repeatedly, so let's give it a go. We explicitly do not allow matching
on names that have already been changed (from a previous udev run, or otherwise), and matching
on unpredictable names (ethX) is discouraged (but not currently disallowed).
We also currently allow:
[Match]
Name=veth0
[Link]
Name=my-name0
SomeOtherSetting=true
Which means that the link file will be applied the first time it is invoked, but
not on subsequent invocations, which may be surprising.
..so make them cry and print a warning if __NR_memfd_create is not
defined. This should make syscall() fail with -ENOSYS, thus trigger a
suitable runtime error-path.
This patch changes the naming scheme for sas disks. The original names used
disk's sas address and lun, the new scheme uses sas address of the
nearest expander (if available) and a phy id of the used connection.
If no expander is used, the phy id of hba phy is used.
Note that names that refer to RAID or other abstract devices are
unchanged.
Name in raid configuration:
hba_pci_address-sas-raid_sas_address-lunY-partZ
Name in expander bare disk configuration:
hba_pci_address-sas-expander_sas_address-phyX-lunY-partZ
Name format without expanders:
hba_pci_address-sas-phyX-lunY-partZ
Signed-off-by: Maurizio Lombardi <mlombard@redhat.com>
This patch add support to specify path cost of the
bridge port to be configured via conf file.
Exampe: conf
file: br.netdev
[NetDev]
Name=br-test
Kind=bridge
file: br.network
[Match]
Name=em1
[Network]
Bridge=br-test
[BridgePort]
Cost=332
bridge link
2: em1 state UP : <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 master
br-test state disabled priority 32 cost 332
We need original socket_fd around otherwise mac_selinux_get_child_mls_label
fails with -EINVAL return code. Also don't call setexeccon twice but rather pass
context value of SELinuxContext option as an extra argument.
This is useful inside of containers or local networks to intrdouce a
stable name of the default gateway host (in case of containers usually
the host, in case of LANs usually local router).
The errors are prefixed with "libxkbcommon" to provide some context,
because they are quite confusing without it. With the prefix, we at
least know where they come from.
If /etc was read only at boot time with an empty /etc/machine-id, the latter
will be mounted as a tmpfs and get reset at each boot. If the system becomes rw
later, this functionality enables to commit in a race-free manner the
transient machine-id to disk.
Generate the file name from ID_PATH plus the rfkill type (wlan,
bluetooth, ...) and ignore the rfkill device name, since it apparently
is not a stable identifier.
Also, ensure that devices disappearing don't result in broken services,
simply exit cleanly.
During upgrades and when transitioning between different systemd
versions in initrd and on the host we have to expect that some
serialization fields are unknown or parse incorrectly. This shouldn't
really be considered an error, hence downgrade the log messages about
it to debug. This way we can still trace it, but it doesn't confuse
users.
This kinda reverts 46849c3f.
This is the IP address of the default route on the link, if present. A
description is printed when available (the manufacturer of the gateway NIC based
on its MAC address).
In the future we should prefer LLDP information over MAC info.
The RFC says to encode an single empty TXT string instead of an empty
TXT array. It also says to treat a zero-length TXT RR as a TXT array
with a single zero-length string.
This is mostly likely the audit socket, and we really should close it
if we cannot make sense of it, since as long as it is open the kernel
might disable the kmsg forwarding of audit msgs, and we should avoid
that, since audit msgs might get completely lost then.
I also downgraded the log message we show a bit, after all things should
really work fine, and we proceed fine with it.
systemd-delta man page promises that multiple types of deltas will be
concatenated if they are listed with a comma as separator. Replace
FOREACH_WORD() with FOREACH_WORD_SEPARATOR() to restore the functionality.
It helps editing units by either creating a drop-in file, like
/etc/systemd/system/my.service.d/override.conf, or by copying the
original unit from /usr/lib/systemd/ to /etc/systemd/ if the --full
option is specified.
It invokes an editor on temporary files related to the unit files and
if the editor exited successfully, then it renames the temporary files
to their original names (e.g. my.service or override.conf) and
daemon-reload is invoked.
If the temporary file is empty the modification is canceled.
See https://bugzilla.redhat.com/show_bug.cgi?id=906824
Also accept '\r' as newline character.
This dropps warnings of the type:
invalid key/value pair in file /usr/lib/udev/rules.d/40-usb-media-players.rules
on line 26, starting at character 25 ('')
The current code would print the character following the first invalid
character.
Given an udev rules-file without a trailing newline we would otherwise print
garbage:
invalid key/value pair in file /usr/lib/udev/rules.d/40-usb-media-players.rules
on line 26, starting at character 25 ('m')
This is now changed to print
invalid key/value pair in file /usr/lib/udev/rules.d/40-usb-media-players.rules
on line 26, starting at character 25 ('')
(still not very good as printing \0 just gives the empty string)
Currently a property in the form of
FOO=bar
is stored as " FOO=bar", i.e. the property name contains a leading space.
That's quite hard to spot.
This patch discards all extra whitespaces but the first one which is required
by libudev's hwdb_add_property.
[zj: modify the check a bit]
https://bugs.freedesktop.org/show_bug.cgi?id=82311
When creating a new mount unit after an event on /proc/self/mountinfo,
check the mount options as well as the fstype to determine if this is a
remote mount that requires network access.
This is an attempt to add it the remote-fs dependencies to a mount unit
if the options change, like when the utab options are picked up after
mountinfo has already been processed. It just adds the remote-fs
dependencies, leaving the local-fs ones in place.
With this change I always get mount units with proper remote-fs
dependencies when mounted with the _netdev option.
Parsing the mount table with libmount races against the mount command,
which will handle the actual mounting before updating utab. This means
the poll event on /proc/self/mountinfo can kick of a reparse in systemd
before the utab information is available.
This change adds in an additional event source using inotify to watch
for changes to utab. It only watches for IN_MOVED_TO events, matching
libmount behavior of always overwriting this file using rename(2).
This does add a second pass through the mount table parsing when utab is
updated.
Let's do this right from the beginning, to prepare ground for udev
messages that most likely want to store list of strings (for device
tags) in messages, and filter on them.
