| Age | Commit message (Collapse) | Author |
|---|
|
The sysroot mount is already taken care of by the add_sysroot_mount function. With this condition left in, we can get something like this:
initrd-root-fs.target.requires
`-- usr.mount -> /run/systemd/generator/usr.mount
in the main system (i.e., not in the initramfs). In the initramfs, the previous condition already kicks in.
|
|
It enhances tunnel(IPIP,GRE, SIT) to aconfigure address as
any.
Fixes #2279
|
|
Fixes:
==1== HEAP SUMMARY:
==1== in use at exit: 67,182 bytes in 91 blocks
==1== total heap usage: 70,485 allocs, 70,394 frees, 42,184,635 bytes
allocated
==1==
==1== 5,742 (696 direct, 5,046 indirect) bytes in 29 blocks are
definitely lost in loss record 4 of 7
==1== at 0x4C2DD9F: realloc (in
/usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==1== by 0x21ADDD: realloc_multiply (alloc-util.h:67)
==1== by 0x21BFB0: strv_push (strv.c:448)
==1== by 0x21C245: strv_consume (strv.c:520)
==1== by 0x21C33C: strv_extend (strv.c:559)
==1== by 0x278AD7: unit_write_drop_in (unit.c:3352)
==1== by 0x278EEB: unit_write_drop_in_private (unit.c:3403)
==1== by 0x190C21: bus_service_set_transient_property
(dbus-service.c:254)
==1== by 0x190DBC: bus_service_set_property (dbus-service.c:284)
==1== by 0x18F00E: bus_unit_set_properties (dbus-unit.c:1226)
==1== by 0x186F6A: transient_unit_from_message (dbus-manager.c:683)
==1== by 0x1872B7: method_start_transient_unit (dbus-manager.c:763)
==1==
==1== LEAK SUMMARY:
==1== definitely lost: 696 bytes in 29 blocks
==1== indirectly lost: 5,046 bytes in 58 blocks
==1== possibly lost: 0 bytes in 0 blocks
==1== still reachable: 61,440 bytes in 4 blocks
==1== suppressed: 0 bytes in 0 blocks
|
|
Prior to this change every journal append causes an ftruncate() for the
sake of inotify propagation of the mmap-based writes.
With this change the notification is deferred up to ~250ms, coalescing
any repeated journal writes during the deferred period into a single
ftruncate(). The ftruncate() call isn't free and doing it on every
append adds unnecessary overhead and latency in the journald event loop.
Introduces journal_file_enable_post_change_timer() which manages a
timer on the provided sd-event instance for scheduling coalesced
ftruncates. The ftruncate() behavior is unchanged unless
journal_file_enable_post_change_timer() is called on the JournalFile.
While not a tremendous improvement, profiling systemd-journald event loop
latencies using instrumentation as introduced by 34b8751 it was observed that
coalescing the ftruncates was low-hanging fruit worth pursuing.
Note orders 12 and 13 shifting left into order 11 and order 6 dipping into
order 5:
Unmodified:
log2(us) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-----------------------------------------------------------
[10685.414572] 0 0 0 0 38 602 61 2 290 60 1643 2554 13 1 4 1 0 0 1
[10690.415114] 0 0 0 0 0 646 54 7 309 44 2073 2148 17 1 3 0 0 0 1
[10695.415509] 0 0 0 0 1 650 73 3 324 37 2071 2270 9 0 0 1 0 1 0
[10700.416297] 0 0 0 0 0 659 50 4 318 38 2111 2152 6 0 1 0 0 1 1
[10705.417136] 0 0 0 0 2 660 48 4 320 38 2129 2146 12 1 1 0 0 1 1
[10710.489114] 0 0 0 0 0 673 38 3 321 37 1925 2339 7 0 0 0 0 1 1
[10715.489613] 0 0 0 0 3 656 64 8 317 48 2365 2007 7 0 0 0 0 0 1
Coalesced:
log2(us) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-----------------------------------------------------------
[ 6169.161360] 0 0 0 1 24 786 54 11 389 24 4192 771 6 4 0 0 1 0 1
[ 6174.161705] 0 0 0 1 18 800 35 6 380 27 3977 893 3 1 0 0 1 0 1
[ 6179.162741] 0 0 0 1 28 768 51 4 391 16 3998 831 5 3 0 0 0 0 2
[ 6184.162856] 0 0 0 0 19 770 60 2 376 26 3795 1004 9 5 1 0 1 0 1
[ 6189.163279] 0 0 0 0 28 761 49 7 372 27 3729 1056 3 2 0 0 1 0 1
[ 6194.164255] 0 0 0 0 25 785 49 7 394 19 3996 908 6 3 2 0 0 0 1
[ 6199.164658] 0 0 0 0 29 797 35 5 389 18 3995 898 3 4 1 1 1 0 1
The remaining high-order delays are a result of the synchronous fsyncs in
systemd-journald, beyond the scope of this commit.
|
|
transaction: downgrade warnings about wanted units which are not found
|
|
Fourteenth DNSSEC PR
|
|
[v4] bus-util: print "systemctl --user" on user service manager
|
|
If a unit was pulled by a Wants= dependency but its unit file was not
present then we logged this as an error.
However Wants= might be used to configure a soft/optional dependency
on another unit, ie. start an optional service only if it's installed
otherwise simply skip it. In this case emitting an error doesn't look
appropriate.
But it's still an error if the optional dependency exists but its
activation fails for any reasons.
|
|
==1== HEAP SUMMARY:
==1== in use at exit: 61,728 bytes in 22 blocks
==1== total heap usage: 258,122 allocs, 258,100 frees, 78,219,628
bytes allocated
==1==
==1== 16 bytes in 1 blocks are definitely lost in loss record 1 of 6
==1== at 0x4C2BBCF: malloc (vg_replace_malloc.c:299)
==1== by 0x1E350E: memdup (alloc-util.c:34)
==1== by 0x135AFB: memdup_multiply (alloc-util.h:74)
==1== by 0x140F97: manager_set_default_rlimits (manager.c:2929)
==1== by 0x1303DA: manager_set_defaults (main.c:737)
==1== by 0x133A02: main (main.c:1718)
==1==
==1== 272 bytes in 17 blocks are definitely lost in loss record 2 of 6
==1== at 0x4C2BBCF: malloc (vg_replace_malloc.c:299)
==1== by 0x1E350E: memdup (alloc-util.c:34)
==1== by 0x135AFB: memdup_multiply (alloc-util.h:74)
==1== by 0x140F97: manager_set_default_rlimits (manager.c:2929)
==1== by 0x1303DA: manager_set_defaults (main.c:737)
==1== by 0x13480D: main (main.c:1828)
==1==
==1== LEAK SUMMARY:
==1== definitely lost: 288 bytes in 18 blocks
==1== indirectly lost: 0 bytes in 0 blocks
==1== possibly lost: 0 bytes in 0 blocks
==1== still reachable: 61,440 bytes in 4 blocks
==1== suppressed: 0 bytes in 0 blocks
==1== Reachable blocks (those to which a pointer was found) are not
shown.
==1== To see them, rerun with: --leak-check=full --show-leak-kinds=all
|
|
When a unit was started with "systemctl --user" and it failed, error
messages is printed as "systemctl status". But it should be "systemctl
--user status".
|
|
Compare errno with zero in a way that tells gcc that
(if the condition is true) errno is positive.
|
|
gcc is confused by the common idiom of
return errno ? -errno : -ESOMETHING
and thinks a positive value may be returned. Replace this condition
with errno > 0 to help gcc and avoid many spurious warnings. I filed
a gcc rfe a long time ago, but it hard to say if it will ever be
implemented [1].
Both conventions were used in the codebase, this change makes things
more consistent. This is a follow up to bcb161b0230f.
[1] https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61846
|
|
We frequently unescape DNS label follwed by IDNA undoing. We now have a function that does that in one step, hence use
it everywhere.
|
|
|
|
This way we become compatible with DNS names with embedded NUL bytes.
|
|
skip first label
|
|
In contrast to ascii_strcasecmp_nn() it takes two character buffers with their individual length. It will then compare
the buffers up the smaller size of the two buffers, and finally the length themselves.
|
|
ascii_strcasecmp_n()
This makes our code compatible with embedded NUL bytes, as we don't care about NUL bytes anymore.
|
|
|
|
|
|
Properly handle RRs that begin with an asterisk label. These are the unexpanded forms of wildcard domains and appear in
NSEC RRs for example. We need to make sure we handle the signatures of these RRs properly, since they mostly are
considered normal RRs, except that the RRSIG labels counter is one off for them, as the asterisk label is always
excluded of the signature.
|
|
Let's determine the source of synthesis once instead of for each RR in the RRset.
|
|
dns_name_startswith() is to dns_name_endswith() as startswith() is to endswith().
|
|
|
|
Closes: #2299
|
|
|
|
|
|
Add machine-id setting
|
|
Allow for overriding all other machine-ids which may be present on
the system using a kernel command line systemd.machine_id or
--machine-id= option.
This is especially useful for network booted systems where the
machine-id needs to be static, or for containers where a specific
machine-id is wanted.
|
|
|
|
sd-event: instrument sd_event_run() for profiling delays
|
|
Set SD_EVENT_PROFILE_DELAYS to activate accounting and periodic logging
of the distribution of delays between sd_event_run() calls.
Time spent in dispatching as well as time spent outside of
sd_event_run() is measured and accounted for. Every 5 seconds a
logarithmic histogram loop iteration delays since 5 seconds previous is
logged.
This is useful in identifying the frequency and magnitude of latencies
affecting the event loop, which should be kept to a minimum.
|
|
Also add a coccinelle receipt to help with such transitions.
|
|
capabilities: added support for ambient capabilities.
|
|
Fix miscalculated buffer size and uses of size-unlimited sprintf()
|
|
The ambient capability tests are only run if the kernel has support for
ambient capabilities.
|
|
This patch adds support for ambient capabilities in service files. The
idea with ambient capabilities is that the execed processes can run with
non-root user and get some inherited capabilities, without having any
need to add the capabilities to the executable file.
You need at least Linux 4.3 to use ambient capabilities. SecureBit
keep-caps is automatically added when you use ambient capabilities and
wish to change the user.
An example system service file might look like this:
[Unit]
Description=Service for testing caps
[Service]
ExecStart=/usr/bin/sleep 10000
User=nobody
AmbientCapabilities=CAP_NET_ADMIN CAP_NET_RAW
After starting the service it has these capabilities:
CapInh: 0000000000003000
CapPrm: 0000000000003000
CapEff: 0000000000003000
CapBnd: 0000003fffffffff
CapAmb: 0000000000003000
|
|
Change the capability bounding set parser and logic so that the bounding
set is kept as a positive set internally. This means that the set
reflects those capabilities that we want to keep instead of drop.
|
|
Now, config_parse_exec() use the userdata as unit and assert on NULL. So
allocate empty unit and use that on config_parse_exec().
|
|
When parse ExecXXX=, specifiers are not resolved in
config_parse_exec(). Finally, the specifiers are set into unit
properties. So, systemctl shows not resolved speicifier on "Process:"
field.
To set the exec properties well, resolve specifiers before parse the
rvale by unit_full_printf();
|
|
journal: normalize priority of logging sources
|
|
function.
Not sure if this results in an exploitable buffer overflow, probably not
since the the int value is likely sanitized somewhere earlier and it's
being put through a bit mask shortly before being used.
|
|
|
|
We wouldn't know how to validate them, since they are the signatures, and hence have no signatures.
|
|
Given how fragile DNS servers are with some DNS types, and given that we really should avoid confusing them with
known-weird lookups, refuse doing lookups for known-obsolete RR types.
|
|
current_feature_level
This is a follow-up for f4461e5641d53f27d6e76e0607bdaa9c0c58c1f6.
|
|
its own
A suggested by Vito Caputo:
https://github.com/systemd/systemd/pull/2289#discussion-diff-49276220
|
|
|
|
DNSSEC
Move detection into a set of new functions, that check whether one specific server can do DNSSEC, whether a server and
a specific transaction can do DNSSEC, or whether a transaction and all its auxiliary transactions could do so.
Also, do these checks both before we acquire additional RRs for the validation (so that we can skip them if the server
doesn't do DNSSEC anyway), and after we acquired them all (to see if any of the lookups changed our opinion about the
servers).
THis also tightens the checks a bit: a server that lacks TCP support is considered incompatible with DNSSEC too.
|
|
This makes it easier to log information about a specific DnsServer object.
|
