thirdparty/systemd.git
16 months agonetwork: ignore requested ipv6 routing policy rule when ipv6 is disabled by sysctl
Yu Watanabe [Tue, 11 Jun 2019 14:29:57 +0000 (23:29 +0900)] 
network: ignore requested ipv6 routing policy rule when ipv6 is disabled by sysctl

16 months agonetwork: ignore requested ipv6 route when ipv6 is disabled by sysctl
Yu Watanabe [Tue, 11 Jun 2019 14:26:11 +0000 (23:26 +0900)] 
network: ignore requested ipv6 route when ipv6 is disabled by sysctl

16 months agonetwork: ignore requested ipv6 addresses when ipv6 is disabled by sysctl
Yu Watanabe [Tue, 11 Jun 2019 14:20:56 +0000 (23:20 +0900)] 
network: ignore requested ipv6 addresses when ipv6 is disabled by sysctl

16 months agoman: fix reference to sd_bus_add_fallback_vtable
Peter A. Bigot [Sun, 9 Jun 2019 17:15:09 +0000 (12:15 -0500)] 
man: fix reference to sd_bus_add_fallback_vtable

16 months agoMerge pull request #12685 from yuwata/network-dhcp-assign-adn-remove-12676
Zbigniew Jędrzejewski-Szmek [Tue, 11 Jun 2019 07:52:26 +0000 (09:52 +0200)] 
Merge pull request #12685 from yuwata/network-dhcp-assign-adn-remove-12676

network: assign new DHCP address before removing old lease address

16 months agosystemd-mount: don't check for non-normalized WHAT for network FS
Karel Zak [Mon, 10 Jun 2019 13:18:47 +0000 (15:18 +0200)] 
systemd-mount: don't check for non-normalized WHAT for network FS

The WHAT string could be whatever for many filesystems. The common
example are network filesystems.

Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=1708996

16 months agoman/systemd-veritysetup-generator: Fix grammar
Theo Ouzhinski [Sat, 8 Jun 2019 21:43:59 +0000 (17:43 -0400)] 
man/systemd-veritysetup-generator: Fix grammar

16 months agonetwork: assign new DHCP address before removing old lease address 12685/head
Yu Watanabe [Fri, 7 Jun 2019 06:31:07 +0000 (15:31 +0900)] 
network: assign new DHCP address before removing old lease address

Closes #12676.

16 months agonetwork: ignore callback calls when link is in failed state
Yu Watanabe [Fri, 7 Jun 2019 06:04:11 +0000 (15:04 +0900)] 
network: ignore callback calls when link is in failed state

16 months agonetwork: drop unnecessary link_enter_failed() calls
Yu Watanabe [Fri, 7 Jun 2019 05:28:09 +0000 (14:28 +0900)] 
network: drop unnecessary link_enter_failed() calls

As the function called soon later anyway.

16 months agonetwork: check earlier the existence of lifetime in lease
Yu Watanabe [Fri, 7 Jun 2019 05:22:13 +0000 (14:22 +0900)] 
network: check earlier the existence of lifetime in lease

16 months agonetwork: make all failures in route configuration fatal
Yu Watanabe [Fri, 7 Jun 2019 05:18:55 +0000 (14:18 +0900)] 
network: make all failures in route configuration fatal

16 months agoMerge pull request #12756 from cdown/uninit
Yu Watanabe [Fri, 7 Jun 2019 06:58:30 +0000 (15:58 +0900)] 
Merge pull request #12756 from cdown/uninit

a couple of correctness fixups

16 months agocgroup: Prevent theoretical nullptr deref in unit mask calculation 12756/head
Chris Down [Fri, 7 Jun 2019 05:28:10 +0000 (06:28 +0100)] 
cgroup: Prevent theoretical nullptr deref in unit mask calculation

16 months agosystemctl: Prevent state_missing from being used uninit
Chris Down [Fri, 7 Jun 2019 05:22:41 +0000 (06:22 +0100)] 
systemctl: Prevent state_missing from being used uninit

16 months agoMerge pull request #12738 from yuwata/network-routing-policy-cleanup
Zbigniew Jędrzejewski-Szmek [Thu, 6 Jun 2019 16:53:20 +0000 (18:53 +0200)] 
Merge pull request #12738 from yuwata/network-routing-policy-cleanup

network: several cleanups for routing policy rule

16 months agoMerge pull request #12511 from ssahani/high-avilability-12050
Zbigniew Jędrzejewski-Szmek [Thu, 6 Jun 2019 16:51:46 +0000 (18:51 +0200)] 
Merge pull request #12511 from ssahani/high-avilability-12050

networkd: add new KeepConfiguration= setting

16 months agotest-network: add tests for KeepConfiguration= 12511/head
Yu Watanabe [Mon, 3 Jun 2019 04:21:13 +0000 (13:21 +0900)] 
test-network: add tests for KeepConfiguration=

16 months agoman: add documentation about KeepConfiguration
Yu Watanabe [Mon, 3 Jun 2019 06:37:25 +0000 (15:37 +0900)] 
man: add documentation about KeepConfiguration

16 months agonetwork: make KeepConfiguration=static drop DHCP addresses and routes
Yu Watanabe [Mon, 3 Jun 2019 03:33:13 +0000 (12:33 +0900)] 
network: make KeepConfiguration=static drop DHCP addresses and routes

Also, KeepConfiguration=dhcp drops static foreign addresses and routes.

16 months agonetwork: add KeepConfiguration=dhcp-on-stop
Yu Watanabe [Mon, 3 Jun 2019 17:05:26 +0000 (02:05 +0900)] 
network: add KeepConfiguration=dhcp-on-stop

The option prevents to drop lease address on stop.
By setting this, we can safely restart networkd.

16 months agonetworkd: add support to keep configuration
Susant Sahani [Mon, 3 Jun 2019 03:31:13 +0000 (12:31 +0900)] 
networkd: add support to keep configuration

16 months agoMerge pull request #12748 from keszybz/modernize-networkd-tests-style
Yu Watanabe [Thu, 6 Jun 2019 13:23:09 +0000 (22:23 +0900)] 
Merge pull request #12748 from keszybz/modernize-networkd-tests-style

Modernize networkd-tests.py style

16 months agoNEWS: add note about the CPUAffinity= changes
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 08:41:46 +0000 (10:41 +0200)] 
NEWS: add note about the CPUAffinity= changes

16 months agoMerge pull request #12508 from keszybz/no-root-checks
Yu Watanabe [Thu, 6 Jun 2019 12:31:19 +0000 (21:31 +0900)] 
Merge pull request #12508 from keszybz/no-root-checks

Drop many root checks

16 months agonetworkd-tests.py: properly escape pluses in regexps 12748/head
Zbigniew Jędrzejewski-Szmek [Thu, 6 Jun 2019 12:08:01 +0000 (14:08 +0200)] 
networkd-tests.py: properly escape pluses in regexps

Python warns that '\+' is an invalid escape. We need either '\\+' or r'\+'.

16 months agonetworkd-tests.py: remove some java-style boilerplate
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 10:08:09 +0000 (12:08 +0200)] 
networkd-tests.py: remove some java-style boilerplate

A class was used to put some utility functions under a namespace. But then this
namespace was inherited into classes, so the namespace split was undone. Let's
just define those functions in the module namespace.

(The Utilities class with a few functions still remain, because of the
unittest-style self.assert* craziness.)

16 months agonetworkd-tests.py: use check_output() instead of unittests' assertEqual()
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 10:03:01 +0000 (12:03 +0200)] 
networkd-tests.py: use check_output() instead of unittests' assertEqual()

It's pretty much the same thing, but more in pytest style, without the
unittest boilerplate.

16 months agonetworkd-tests.py: add helpers to common subprocess call patterns
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 09:43:28 +0000 (11:43 +0200)] 
networkd-tests.py: add helpers to common subprocess call patterns

For normal shell calls, python syntax is quite verbose. We don't need to punish
ourselves like that. In some places we would fork a shell to do argument splitting
for us. We know that our arguments can be safely split on whitespace, so let's do
that ourselves in all cases, without forking a shell.

We also expect command output to be valid text, so let's always set
universal_newlines=True.

This makes things shorter and easier to read. Development is also easier because
it's possible to paste many of the commands directly to/from a shell.

16 months agonetworkd-tests.py: remove unused import
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 10:42:29 +0000 (12:42 +0200)] 
networkd-tests.py: remove unused import

16 months agonetwork: check stored object is euivalent to what we want to remove 12738/head
Yu Watanabe [Wed, 5 Jun 2019 15:42:05 +0000 (00:42 +0900)] 
network: check stored object is euivalent to what we want to remove

When object A is stored in Manager::rules and B is in ::rules_foreign,
and compare function for the object cannot distinguish them,
then freeing A causes B to be removed from rules_foreign or vice versa.

Hopefully fixes #12731.

16 months agoDrop no longer needed halt-local.service.in
Michael Biebl [Wed, 5 Jun 2019 11:35:11 +0000 (13:35 +0200)] 
Drop no longer needed halt-local.service.in

Follow-up for commit 44508946534eee032927c263b79464832656dd6e.

16 months agoMerge pull request #12744 from yuwata/test-network-stability-dhcp-tests
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 12:31:24 +0000 (14:31 +0200)] 
Merge pull request #12744 from yuwata/test-network-stability-dhcp-tests

test-network: improve stability of DHCP client tests

16 months agotest-network: do not directly compare two results of "ip address" 12744/head
Yu Watanabe [Tue, 4 Jun 2019 18:57:21 +0000 (03:57 +0900)] 
test-network: do not directly compare two results of "ip address"

As it contains lifetime of the address.

16 months agoMerge pull request #11778 from anitazha/rfe_11654_dbus
Chris Down [Wed, 5 Jun 2019 09:02:00 +0000 (10:02 +0100)] 
Merge pull request #11778 from anitazha/rfe_11654_dbus

core: add ExecStartXYZEx= with dbus support for executable prefixes

16 months agoudev: propagate errors in udev_event_execute_rules()
Yu Watanabe [Tue, 4 Jun 2019 18:09:20 +0000 (03:09 +0900)] 
udev: propagate errors in udev_event_execute_rules()

And do not set initialized flag to the device.

16 months agoudevadm trigger: log errors and return first failure 12508/head
Zbigniew Jędrzejewski-Szmek [Wed, 5 Jun 2019 07:54:54 +0000 (09:54 +0200)] 
udevadm trigger: log errors and return first failure

When udevadm trigger is called, the list of devices to trigger is always
generated through enumeration, and devices can come and go, so we should not
treat -ENOENT as a failure. But other types of failure should be logged.
It seems they were logged until baa30fbc2c04b23209d0b8fb3c86cd15ef9ea81a.

Also, return the first error. (I'm not sure if there are other failure modes
which we want to ignore. If they are, they'll need to be whitelisted like
-ENOENT.).

16 months agotest-network: do not allow addresses in tentative state
Yu Watanabe [Tue, 4 Jun 2019 18:55:32 +0000 (03:55 +0900)] 
test-network: do not allow addresses in tentative state

16 months agoman: mention quoting in description of Parameters=
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 11:44:46 +0000 (13:44 +0200)] 
man: mention quoting in description of Parameters=

Also add a short example how to translate nspawn command line to .nspawn settings.

Fixes #12703.

16 months agoMerge pull request #12741 from keszybz/bpf-firewall-warning
Yu Watanabe [Tue, 4 Jun 2019 19:01:44 +0000 (04:01 +0900)] 
Merge pull request #12741 from keszybz/bpf-firewall-warning

Silence BPF firewall warning when not useful

16 months agoMerge pull request #12737 from keszybz/meson-man-nspawn
Yu Watanabe [Tue, 4 Jun 2019 17:45:39 +0000 (02:45 +0900)] 
Merge pull request #12737 from keszybz/meson-man-nspawn

meson: escape dots in man/man helper

16 months agoMerge pull request #12661 from mrc0mmand/debug-journalctl-flush
Yu Watanabe [Tue, 4 Jun 2019 15:46:29 +0000 (00:46 +0900)] 
Merge pull request #12661 from mrc0mmand/debug-journalctl-flush

journal: disable varlink timeout for journalctl --flush or friends

16 months agoman/man: refer to the right build directory in the helpers 12737/head
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 15:33:59 +0000 (17:33 +0200)] 
man/man: refer to the right build directory in the helpers

I have no idea how this one slipped through.

16 months agoSilence warning about BPF firewall in containers 12741/head
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 15:22:12 +0000 (17:22 +0200)] 
Silence warning about BPF firewall in containers

We'd get a warning on every nspawn invocation:
dev-hugepages.mount: unit configures an IP firewall, but the local system does not support BPF/cgroup firewalling.
(This warning is only shown for the first unit using IP firewalling.)

Before the previous commit, I'd generally get a warning about systemd-udev.service, even though that service is
not started in containers. But are still many other units which that declare a
firewall, which is currently unsupported in containers. Let's stop warning
about this.

The warning is still emitted e.g. if legacy cgroups are used. This is something
that can be configured, so it makes more sense to emit the warning.

16 months agoMove warning about unsupported BPF firewall right before the firewall would be created
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 13:01:27 +0000 (15:01 +0200)] 
Move warning about unsupported BPF firewall right before the firewall would be created

There's no need to warn about the firewall when parsing, because the unit might
not be started at all. Let's warn only when we're actually preparing to start
the firewall.

This changes behaviour:
- the warning is printed just once for all unit types, and not once
  for normal units and once for transient units.
- on repeat warnings, the message is not printed at all. There's already
  detailed debug info from bpf_firewall_compile(), so we don't need to repeat
  ourselves.
- when we are not root, let's say precisely that, not "lack of necessary privileges"
  and "the local system does not support BPF/cgroup firewalling".

Fixes #12673.

16 months agojournalctl: make 'journalctl --flush' or friends not fail with varlink timeout 12661/head
Yu Watanabe [Fri, 24 May 2019 13:32:08 +0000 (22:32 +0900)] 
journalctl: make 'journalctl --flush' or friends not fail with varlink timeout

Closes #12570.

16 months agomeson: escape dots in man/man helper
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 10:56:14 +0000 (12:56 +0200)] 
meson: escape dots in man/man helper

It was impossible to view systemd.nspawn(5), because systemd-nspawn(1) was matched
also, and happened to be earlier in the list. The solution in this patch is pretty
crude, but is should be enough for our purposes, since we don't have any regexp
special characters in man page names except for the dot.

16 months agoMerge pull request #12733 from keszybz/sd-bus-error-reporting
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 10:01:52 +0000 (12:01 +0200)] 
Merge pull request #12733 from keszybz/sd-bus-error-reporting

Bus error reporting improvements

16 months agoMerge pull request #12734 from keszybz/cpu-set-util-on-i386
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 10:01:20 +0000 (12:01 +0200)] 
Merge pull request #12734 from keszybz/cpu-set-util-on-i386

Fixes for test-cpu-set-util on i386

16 months agobash-completion: support "networkctl delete"
Yu Watanabe [Tue, 4 Jun 2019 08:13:07 +0000 (17:13 +0900)] 
bash-completion: support "networkctl delete"

16 months agojournalctl: mention --smart-relinquish-var in log message
Yu Watanabe [Mon, 27 May 2019 20:17:27 +0000 (05:17 +0900)] 
journalctl: mention --smart-relinquish-var in log message

16 months agojournalctl: fix error cause in log message
Yu Watanabe [Fri, 24 May 2019 13:22:21 +0000 (22:22 +0900)] 
journalctl: fix error cause in log message

If varlink_call() returns negative errno, then `error` is null.

16 months agotest-cpu-set-util: fix allocation size check on i386 12734/head
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 07:40:38 +0000 (09:40 +0200)] 
test-cpu-set-util: fix allocation size check on i386

We get just 28 bytes not 32 as on 64-bit architectures (__cpu_set_t is 4 bytes,
we need at least 26, so 28 satisfies the constraints).

16 months agotest-cpu-set-util: fix comparison for allocation size
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 07:19:04 +0000 (09:19 +0200)] 
test-cpu-set-util: fix comparison for allocation size

On i386, __cpu_mask is 4 bytes, so we'd check if c.allocated >= 0, and
gcc would warn about a bogus comparison. Let's round up.

Fixes #12726.

16 months agonetwork: remove unused argument in routing_policy_rule_configure()
Yu Watanabe [Tue, 4 Jun 2019 05:40:21 +0000 (14:40 +0900)] 
network: remove unused argument in routing_policy_rule_configure()

16 months agonetwork: fix double free on error path
Yu Watanabe [Tue, 4 Jun 2019 05:25:09 +0000 (14:25 +0900)] 
network: fix double free on error path

16 months agonetwork: fix UBSAN issue
Yu Watanabe [Mon, 3 Jun 2019 14:29:07 +0000 (23:29 +0900)] 
network: fix UBSAN issue

Closes #12724.

16 months agonetworkctl: re-cast value stored in Set with INT_TO_PTR()
Yu Watanabe [Tue, 4 Jun 2019 05:00:51 +0000 (14:00 +0900)] 
networkctl: re-cast value stored in Set with INT_TO_PTR()

16 months agoshared/generator: voidify mkdir_parents() call
Zbigniew Jędrzejewski-Szmek [Tue, 4 Jun 2019 06:57:54 +0000 (08:57 +0200)] 
shared/generator: voidify mkdir_parents() call

To appease coverity: CID#1401752.

16 months agobusctl: always prefix bus error message with local string 12733/head
Zbigniew Jędrzejewski-Szmek [Fri, 31 May 2019 14:21:56 +0000 (16:21 +0200)] 
busctl: always prefix bus error message with local string

The message we get from the bus can be pretty bare-bones, and the user might be
confused which operation failed. Since we don't control the other side, let's
prefix the remote message with an informative prefix.

16 months agosd-bus: improve bus error message for uknown properties
Zbigniew Jędrzejewski-Szmek [Fri, 31 May 2019 13:33:46 +0000 (15:33 +0200)] 
sd-bus: improve bus error message for uknown properties

"Unknown interface or property." →
"Unknown interface org.freedesktop.network1.Link or property BitRates."
(I don't think the quotes are necessary. Dbus names have pretty strict rules.)

16 months agonetworkctl: log quietly if we fail to get bit rates
Zbigniew Jędrzejewski-Szmek [Fri, 31 May 2019 14:08:31 +0000 (16:08 +0200)] 
networkctl: log quietly if we fail to get bit rates

If we query older networked, they will be missing. This is not an error.
Also, make the error more informative, because the bus error itself might
be just "Unknown interface or property" which is not enough to understand
what is going on.

16 months agocpu-set-util: use %d-%d format in cpu_set_to_range_string() only for actual ranges
Michal Sekletar [Mon, 3 Jun 2019 08:12:35 +0000 (10:12 +0200)] 
cpu-set-util: use %d-%d format in  cpu_set_to_range_string() only for actual ranges

16 months agotest-sleep: also print parsed modes and states
Zbigniew Jędrzejewski-Szmek [Wed, 29 May 2019 14:08:47 +0000 (16:08 +0200)] 
test-sleep: also print parsed modes and states

This allows the parser to be verified against on-disk config.

16 months agoMerge pull request #12708 from yuwata/test-network-networkctl-tests
Yu Watanabe [Mon, 3 Jun 2019 18:58:47 +0000 (03:58 +0900)] 
Merge pull request #12708 from yuwata/test-network-networkctl-tests

udevadm: add --wait-for-initialization option and use it in systemd-networkd-tests.py

16 months agotest-network: disable SystemCallFilter= when *SAN is enabled 12708/head
Yu Watanabe [Sat, 1 Jun 2019 06:55:16 +0000 (15:55 +0900)] 
test-network: disable SystemCallFilter= when *SAN is enabled

16 months agotest-network: add test for MTU field of 'networkctl status'
Yu Watanabe [Sat, 1 Jun 2019 06:44:04 +0000 (15:44 +0900)] 
test-network: add test for MTU field of 'networkctl status'

16 months agotest-network: add tests for "networkctl delete"
Yu Watanabe [Fri, 31 May 2019 07:11:53 +0000 (16:11 +0900)] 
test-network: add tests for "networkctl delete"

16 months agotest-network: move networkctl related tests to NetworkctlTests class
Yu Watanabe [Sat, 1 Jun 2019 01:29:08 +0000 (10:29 +0900)] 
test-network: move networkctl related tests to NetworkctlTests class

16 months agoudevadm: add --wait-for-initialization option to "udevadm info"
Yu Watanabe [Sat, 1 Jun 2019 16:02:09 +0000 (01:02 +0900)] 
udevadm: add --wait-for-initialization option to "udevadm info"

16 months agoutil: make device_wait_for_initialization() optionally takes timeout value
Yu Watanabe [Sat, 1 Jun 2019 16:01:31 +0000 (01:01 +0900)] 
util: make device_wait_for_initialization() optionally takes timeout value

16 months agoutil: make argument "subsystem" in device_wait_for_initialization() optional
Yu Watanabe [Sat, 1 Jun 2019 15:44:02 +0000 (00:44 +0900)] 
util: make argument "subsystem" in device_wait_for_initialization() optional

16 months agoMerge pull request #12619 from zachsmith/refactor-parse-sleep-config
Zbigniew Jędrzejewski-Szmek [Mon, 3 Jun 2019 15:30:05 +0000 (17:30 +0200)] 
Merge pull request #12619 from zachsmith/refactor-parse-sleep-config

systemd-sleep: refactor sleep conf parsing

16 months agosd-boot: Remove unnecessary call to GetEventLog
Joe Richey [Fri, 31 May 2019 18:24:20 +0000 (11:24 -0700)] 
sd-boot: Remove unnecessary call to GetEventLog

This PR reverts #4302 and part of #7568, as the call to GetEventLog
isn't necessary to get the logs from the OS runtime.

This has already been reverted in shim, see:
    https://github.com/rhboot/shim/commit/fd7c3bd920ba39082cb7c619afb7203d150a4cd3

The reasoning for the change that added this is wrong, explained below:

> According to TCG EFI Protocol Specification for TPM 2.0 family,
> all events generated after the invocation of EFI_TCG2_GET_EVENT_LOG
> shall be stored in an instance of an EFI_CONFIGURATION_TABLE aka
> EFI TCG 2.0 final events table.

This is correct and a valid reading of the TCG spec.

> Hence, it is necessary to trigger the internal switch through calling
> get_event_log() in order to allow to retrieve the logs from OS runtime.

However, this reasoning is incorrect. The "EFI TCG 2.0 final events
table" exists to allow the OS Runtime to get TCG events created during
or after the call to ExitBootServices. If systemd-boot didn't call
trigger_tcg2_final_events_table(), the call to HashLogExtendEvent will
write the event to the normal events table (i.e. the same place the
PCR 0 - 7 related events are written). Then the OS can obtain these
systemd-boot events in the same way it obtains normal events, via a
call to GetEventLog().

Tested on amd64 w/ TPM2 using the newest Linux kernel patches that
get the TPM2 formatted log:
    https://www.lkml.org/lkml/2019/5/20/1590
I confirmed that this change still lets the OS runtime obtain the log
entries created by systemd-boot.

Signed-off-by: Joe Richey <joerichey@google.com>
16 months agoMerge pull request #12666 from yuwata/networkctl-show-stats
Zbigniew Jędrzejewski-Szmek [Mon, 3 Jun 2019 13:38:49 +0000 (15:38 +0200)] 
Merge pull request #12666 from yuwata/networkctl-show-stats

networkctl: show stats

16 months agoMerge pull request #12700 from yuwata/udev-propagate-critical-errors
Zbigniew Jędrzejewski-Szmek [Mon, 3 Jun 2019 13:29:40 +0000 (15:29 +0200)] 
Merge pull request #12700 from yuwata/udev-propagate-critical-errors

udev: modernize code and propagate critical errors

16 months agoexecute: dump CPUAffinity as a range string instead of a list of CPUs
Michal Sekletar [Fri, 31 May 2019 16:02:20 +0000 (18:02 +0200)] 
execute: dump CPUAffinity as a range string instead of a list of CPUs

We do this already when printing the property in systemctl so be
consistent and do the same for systemd-analyze dump.

16 months agonetworkctl: drop unnecessary use of table_add_cell_full() 12666/head
Yu Watanabe [Mon, 3 Jun 2019 07:04:16 +0000 (16:04 +0900)] 
networkctl: drop unnecessary use of table_add_cell_full()

16 months agoudev: check formatting of attribute or value earlier 12700/head
Yu Watanabe [Tue, 30 Apr 2019 20:22:19 +0000 (22:22 +0200)] 
udev: check formatting of attribute or value earlier

16 months agoudev: evaluate formatting in RUN= key earlier
Yu Watanabe [Mon, 29 Apr 2019 20:24:25 +0000 (22:24 +0200)] 
udev: evaluate formatting in RUN= key earlier

Closes #12291.

16 months agoudev: use delete_trailing_chars() at one more place
Yu Watanabe [Tue, 30 Apr 2019 19:01:44 +0000 (21:01 +0200)] 
udev: use delete_trailing_chars() at one more place

16 months agoudev: replace xsprintf() + strpcpy() by strpcpyf()
Yu Watanabe [Tue, 30 Apr 2019 18:35:08 +0000 (20:35 +0200)] 
udev: replace xsprintf() + strpcpy() by strpcpyf()

16 months agoudev: make parser for attribute of $result harder
Yu Watanabe [Tue, 30 Apr 2019 18:30:49 +0000 (20:30 +0200)] 
udev: make parser for attribute of $result harder

16 months agoudev: split udev_event_apply_format() into small pieces
Yu Watanabe [Tue, 30 Apr 2019 18:09:49 +0000 (20:09 +0200)] 
udev: split udev_event_apply_format() into small pieces

Also, this makes the function refuse invalid formatting and
propagate error in substituing variable.

16 months agoudev: comment that $tempnode and $sysfs formats are deprecated
Yu Watanabe [Tue, 30 Apr 2019 17:23:44 +0000 (19:23 +0200)] 
udev: comment that $tempnode and $sysfs formats are deprecated

16 months agoudev: pass format type to udev_event_subst_format()
Yu Watanabe [Tue, 30 Apr 2019 17:18:33 +0000 (19:18 +0200)] 
udev: pass format type to udev_event_subst_format()

16 months agoudev: rename enum subst_type -> FormatSubstitutionType
Yu Watanabe [Tue, 30 Apr 2019 17:13:58 +0000 (19:13 +0200)] 
udev: rename enum subst_type -> FormatSubstitutionType

16 months agoudev: modernize udev-rules.c
Yu Watanabe [Wed, 24 Apr 2019 23:21:11 +0000 (01:21 +0200)] 
udev: modernize udev-rules.c

This does the following:
- rename enum udev_builtin_cmd -> UdevBuiltinCmd
- rename struct udev_builtin -> UdevBuiltin
- move type definitions to udev-rules.h
- move prototypes of functions defined in udev-rules.c to udev-rules.h
- drop to use strbuf
- propagate critical errors in applying rules,
- drop limitation for number of tokens per line.

16 months agonetworkctl: show link bit rates
Yu Watanabe [Sat, 25 May 2019 15:22:05 +0000 (00:22 +0900)] 
networkctl: show link bit rates

16 months agonetwork: monitor link bit rates
Yu Watanabe [Fri, 24 May 2019 20:08:13 +0000 (05:08 +0900)] 
network: monitor link bit rates

16 months agobash-completion: support --stats/-s option of networkctl
Yu Watanabe [Fri, 24 May 2019 15:40:38 +0000 (00:40 +0900)] 
bash-completion: support --stats/-s option of networkctl

16 months agonetworkctl: optionally show link statistics
Yu Watanabe [Fri, 24 May 2019 15:37:48 +0000 (00:37 +0900)] 
networkctl: optionally show link statistics

16 months agonetlink: set attribute size of IFLA_STATS and IFLA_STATS64
Yu Watanabe [Fri, 24 May 2019 15:37:00 +0000 (00:37 +0900)] 
netlink: set attribute size of IFLA_STATS and IFLA_STATS64

16 months agonetworkctl: shorten Minimum/Maximum to min/max
Zbigniew Jędrzejewski-Szmek [Fri, 31 May 2019 08:56:45 +0000 (10:56 +0200)] 
networkctl: shorten Minimum/Maximum to min/max

This makes the output slightly "lighter" visually. min/max are well-established
forms, so there's no risk of misunderstanding. Also, not using title-case looks
a bit better.

Follow-up for c06ff86e25db4fdb4e962baf812ed28d8d0c1050.

$ build/networkctl status '*'|grep MTU
                   MTU: 65536
                   MTU: 1500 (min: 68, max: 9000)
                   MTU: 1500 (max: 2048)
                   MTU: 1500 (min: 256, max: 2304)
                   MTU: 1500 (min: 68, max: 65535)
                   MTU: 1500 (min: 68, max: 65521)
                   MTU: 1500 (min: 68, max: 65535)
                   MTU: 1500 (max: 65535)
                   MTU: 1360 (min: 68, max: 65535)

16 months agodbus-execute: make transfer of CPUAffinity endian safe (#12711)
Michal Sekletar [Fri, 31 May 2019 13:23:23 +0000 (15:23 +0200)] 
dbus-execute: make transfer of CPUAffinity endian safe (#12711)

We store the affinity mask in the native endian. However, over D-Bus we
must transfer the mask in little endian byte order.

This is the second part of c367f996f5f091a63f812f0140b304c649be77fc.

16 months agocoredump: fixed bug - some coredump temp files could be lost
Krayushkin Konstantin [Fri, 31 May 2019 08:16:17 +0000 (11:16 +0300)] 
coredump: fixed bug - some coredump temp files could be lost

If the machine was suddenly shutted down (hard reboot for example) while
processing core dump, temp files created manually (not with a O_TEMPFILE flag)
stay in the system. After reboot systemd-coredump treat them as usual files, so
they wouldn't be rotated and shall pollute the filesystem.

Solution is to simply add those temp files to systemd-tmpfiles configs.

16 months agoMerge pull request #12707 from yuwata/test-network-lsan-options
Zbigniew Jędrzejewski-Szmek [Fri, 31 May 2019 09:38:18 +0000 (11:38 +0200)] 
Merge pull request #12707 from yuwata/test-network-lsan-options

test-network: add --lsan-options option

16 months agonetworkctl: do not show zero maximum MTU
Yu Watanabe [Thu, 30 May 2019 18:30:31 +0000 (03:30 +0900)] 
networkctl: do not show zero maximum MTU

16 months agotest-network: show systemd-networkd.service 12707/head
Yu Watanabe [Fri, 31 May 2019 04:54:08 +0000 (13:54 +0900)] 
test-network: show systemd-networkd.service

As there may exist some other drop-in configs.

16 months agocore: add ExecStartXYZEx= with dbus support for executable prefixes 11778/head
Anita Zhang [Wed, 20 Feb 2019 22:53:58 +0000 (14:53 -0800)] 
core: add ExecStartXYZEx= with dbus support for executable prefixes

Closes #11654