Lennart Poettering
9a43fc6a2a
Merge pull request #11827 from keszybz/pkgconfig-variables
...
Allow overriding pkgconfig prefixes
2019-02-26 18:58:03 +01:00
Lennart Poettering
88a56c7670
Merge pull request #11357 from GiacintoCifelli/dbus_labels
...
sd-bus: add methods and signals parameter names
2019-02-26 17:22:04 +01:00
Lennart Poettering
8c28360a21
Merge pull request #11823 from keszybz/more-fuzz-coverage
...
More fuzz coverage
2019-02-26 17:21:32 +01:00
Giacinto Cifelli
856ad2a86b
sd-bus: add methods and signals parameter names. Fixes : #1564
2019-02-26 12:55:02 +01:00
Lennart Poettering
a81a4b3b24
Merge pull request #11822 from yuwata/fuzz-udev-database
...
fuzz: add fuzzer for udev database
2019-02-26 10:26:38 +01:00
Anita Zhang
e51237253e
core: consider non-SERVICE_EXEC_START commands for EXIT_CLEAN_COMMAND
...
When there are multiple ExecStop= statements, the next command would continue
to run even after TimeoutStopSec= is up and sends SIGTERM. This is because,
unless Type= is oneshot, the exit code/status would evaluate to SERVICE_SUCCESS
in service_sigchld_event()'s call to is_clean_exit(). This success indicates
following commands would continue running until the end of the list
is reached, or another timeout is hit and SIGKILL is sent.
Since long running processes should not be invoked in non-SERVICE_EXEC_START
commands, consider them for EXIT_CLEAN_COMMAND instead of EXIT_CLEAN_DAEMON.
Passing EXIT_CLEAN_COMMAND to is_clean_exit() evaluates the SIGTERM exit
code/status to failure and will stop execution after the first timeout is hit.
Fixes #11431
2019-02-26 10:18:39 +01:00
Zbigniew Jędrzejewski-Szmek
453823a290
Merge pull request #11780 from yuwata/fix-4211
...
network: skip .network files earlier when conditions do not match system environment
2019-02-26 08:31:32 +01:00
Yu Watanabe
10f6188b67
fuzz: add a sample for fuzz-udev-database
2019-02-26 13:28:30 +09:00
Yu Watanabe
4c9024c2cf
fuzz: add fuzzer for udev database
2019-02-26 13:10:31 +09:00
Yu Watanabe
b07d0f2a85
sd-device: split device_read_db_internal() into two part
...
The new device_read_db_internal_filename() will be used by a fuzzer.
2019-02-26 12:55:43 +09:00
Ryan Gonzalez
c20db38875
cryptsetup: Treat key file errors as a failed password attempt
...
6f177c7dc0
caused key file errors to immediately fail, which would make it hard to correct an issue due to e.g. a crypttab typo or a damaged key file.
Closes #11723 .
2019-02-26 10:48:07 +09:00
Yu Watanabe
102bc043bb
network: assign Network::manager when it is listed to the manager object.
...
Now Network::manager is not necessary during parsing configs.
C.f. 838b2f7a30
.
2019-02-26 10:06:27 +09:00
Yu Watanabe
cebe12571d
network: make resolving NetDev names delayed and moved to network_verify()
...
And before resolving NetDev names, check conditions in .network,
and if they do not match the system environment, drop the network
unit earlier.
Fixes #4211 .
2019-02-26 10:06:23 +09:00
Yu Watanabe
a6779fecd9
network: add debug log when conditions do not match system environment
2019-02-26 09:58:33 +09:00
Lennart Poettering
510dc4da13
Merge pull request #11767 from yuwata/network-bridge-enslaved
...
network: extend 'enslaved' state to bridge slave
2019-02-25 19:29:30 +01:00
Michael Olbrich
416d7d4648
meson: declare version.h as dependency for systemd
...
This is a followup to #11815 and adds the last missing dependency.
With this #11565 is hopefully really fixed.
2019-02-25 15:55:38 +01:00
Zbigniew Jędrzejewski-Szmek
4832ce7eec
fuzz-unit-file: also run manager_dump()
...
This should increase coverage a bit.
2019-02-25 15:46:23 +01:00
Zbigniew Jędrzejewski-Szmek
4c4589227e
fuzz-unit-file: add some directives for stuff coverage reports as not covered
...
Some of those directives appear in the corpus, but without arguments, so maybe
the fuzzing libraries can't trigger the right cases. Let's help them.
2019-02-25 15:46:23 +01:00
Lennart Poettering
c55a447ab4
Merge pull request #11798 from keszybz/mem-sanitizer-fix
...
meson: make sure preprocesor warnings are not treated as errors
2019-02-25 13:50:56 +01:00
Zbigniew Jędrzejewski-Szmek
60722ad778
meson: declare version.h as dep for various targets that include build.h
...
Should fix #11565 .
2019-02-25 10:41:41 +01:00
Zbigniew Jędrzejewski-Szmek
adbdcfbe63
test-json: use standard test intro
2019-02-25 10:07:18 +01:00
Zbigniew Jędrzejewski-Szmek
b6cda3ec4d
test-json: avoid deep stack recursion under msan
2019-02-25 10:07:18 +01:00
Zbigniew Jędrzejewski-Szmek
9003da2963
test-mountpoint-util: unpoison string allocated by sscanf %ms
2019-02-25 10:07:18 +01:00
Zbigniew Jędrzejewski-Szmek
c322f379e6
Add wrapper for __msan_unpoinson() to reduce #ifdeffery
...
This isn't really necessary for the subsequent commit, but I expect that we'll
need to unpoison more often once we turn on msan in CI, so I think think this
change makes sense in the long run.
2019-02-25 10:07:18 +01:00
Zbigniew Jędrzejewski-Szmek
3b8951c1dc
meson: make sure preprocesor warnings are not treated as errors
...
Clang includes -W#warning in -Werror, so the #warning used for msan would
be an error.
v2:
- use -Wno-error=... so that the warning is still emitted, but not as an error.
2019-02-25 10:06:47 +01:00
Zbigniew Jędrzejewski-Szmek
d05da14f32
Merge pull request #11796 from yuwata/fuzz-link-parser
...
fuzz: add fuzzer for .link files
2019-02-25 09:55:02 +01:00
Yu Watanabe
051f39c227
tools: check all directives even if it detects non-updated files
2019-02-25 12:41:11 +09:00
Yu Watanabe
810d43a8e8
tools: update check-directives.sh to support fuzz-link-parser
2019-02-25 12:41:11 +09:00
Yu Watanabe
eac31d708f
fuzz: add directives.link and 99-default.link for fuzz-link-parser
...
Also adds several reproducers of errors fixed in earlier commits.
2019-02-25 12:40:42 +09:00
Yu Watanabe
a378400b3f
fuzz: add fuzzer for parsing .link files
...
This also renames load_link() to link_load_one()
2019-02-25 12:35:51 +09:00
Yu Watanabe
7cd1f60df0
fuzz: use fflush() and drop unnecessary rewind()
2019-02-25 12:35:51 +09:00
Yu Watanabe
84fb56d396
udev/ethtool: fix error detection of ethtool_link_mode_bit_from_string()
2019-02-25 12:35:40 +09:00
Yu Watanabe
391f6bc1db
udev: fix memleak in conditions for .link file
2019-02-25 11:57:14 +09:00
Yu Watanabe
176d9c0e11
udev/net: drop .link files earlier when their conditions do not match system environment
2019-02-25 11:54:50 +09:00
Yu Watanabe
6cdab9f17f
udev/net: use structured initializer at one more place
2019-02-25 11:53:57 +09:00
Yu Watanabe
79a60834e2
udev/net: use size_t for index at one more place
2019-02-25 11:51:32 +09:00
Yu Watanabe
e8a42907ed
udev/net: ignore errors in loading .link files but warn about that
2019-02-25 11:50:57 +09:00
Yu Watanabe
f2d251cdb8
udev/net: shorten load_link() a little bit
2019-02-25 11:48:39 +09:00
Yu Watanabe
c6b3370ab2
udev: drop unused Ethernet section
2019-02-25 11:45:34 +09:00
Yu Watanabe
7d4ea8f91c
man: update explanation about operational state of network interfaces
2019-02-25 09:34:14 +09:00
Yu Watanabe
85fc09c97a
man: mention that LinkLocalAddressing= is disabled by default when Bridge= is set
2019-02-25 09:34:14 +09:00
Yu Watanabe
6609924c32
test-network: add more tests for IgnoreCarrierLoss=
...
Suggested by @amishmm in #9262 .
2019-02-25 09:34:14 +09:00
Yu Watanabe
2be6c5d2ec
test-network: add more tests for Bridge=
2019-02-25 09:34:14 +09:00
Yu Watanabe
45e11abfff
network: make bridge master also follow operstates of slave interfaces
...
If one of bridge slaves is in off, no-carrier, or dormant, then
operstate of the bridge master is set to degraded.
2019-02-25 09:34:14 +09:00
Yu Watanabe
34bf3c0051
network: extend 'enslaved' state to bridge slave interfaces
...
Currently, the interface's operstate is set to 'enslaved' only when
it is managed by networkd.
2019-02-25 09:34:14 +09:00
Yu Watanabe
8f1918016f
network: disable LinkLocalAddressing= and IPv6AcceptRA= on bridge slaves by default
...
If they are not explicitly configured, then let's disable these
functionality on bridge slaves.
2019-02-25 09:34:14 +09:00
Yu Watanabe
b102cdca6b
network: do not disable dynamic addressing for bridge slaves
...
This effectively reverts 5971cb9de9
and
2b00a4e03d
.
Usually, it is not necessary to assign addresses to bridge slaves,
but such functionality is supported by kernel. If users explicitly
request such configuration, networkd should support that.
2019-02-25 09:34:14 +09:00
Yu Watanabe
033295c188
network: fix invalid memory access
...
This fixes a bug introduced by 959f65d32e
.
2019-02-25 09:34:14 +09:00
Frantisek Sumsal
f350965b15
Merge pull request #11808 from yuwata/fix-11806
...
test-network: testing with two bond slaves
2019-02-24 21:25:34 +01:00
Yu Watanabe
cc3e488c7d
test-network: testing with two bond slaves
2019-02-23 19:43:25 +09:00