Before 12001b1bf0, write() is required for
if Type=exec. However, with the previous commit, now write() is also used
for sending handoff timestamp. Let's allow write() if necessary.
Fixes a regression caused by 12001b1bf0.
Fixes#33299.
Before 12001b1bf0, the timestamp is sent
with write(), but the commit made the timestamp sent by send(), and
causes regressin #33299.
Note the invocation will still fail if write() is filtered by seccomp.
But, that is an old issue since Type=exec is introduced
(5686391b00).
Partially fixes a regression caused by 12001b1bf0.
Partially fixes#33299.
Similar to the previous commit, but for preventing from removing static
routes on receiving RA with zero lifetime.
Fixes a regresson caused by 479d3e1994.
Fixes#33346.
We have already ignored conflicting address configurations requested by
NDisc protocol. See ndisc_request_address().
Let's follow the same rule for routes. That is, if there are conflicting
static routes configured or requested, do not override them by NDisc.
Also, swap the order of checking existing route and existing request.
Fixes a regression caused by 972f1d17ab.
Prompted by #33346.
- introduce or rename usual enum values _MAX and _INVALID,
- introduce and use string table lookup functions,
- split out implementation of get_color_mode() to _impl(),
- add tests for get_color_mode().
And drop spurious assertion.
Fortunately, the previous logic worked, as /run/systemd/unit-private-tmp
is ordered earlier than /tmp or /var/tmp. But, let's ensure the tmpfs
mounted earlier to make the logic clearer.
Follow-up for 0e551b04ef.
This also
- merges basic/linux and shared/linux,
- moves BPF_JUMP_A() to basic/missing_bpf.h,
- copies from usrspace kernel headers directory generated by 'make headers',
rather than copying from kernel tree,
- copies const.h into our tree to reduce change in ethtool.h,
- copies auto_fs.h into our tree to reduce change in auto_dev-ioctl.h.
This reverts commit f5bdecba08.
Some kmsg sent before sleep may be received by systemd-journald after
sleep. In that case, map_clock_usec() does not provide correct
timestamp.
So, _SOURCE_MONOTONIC_TIMESTAMP field is anyway unreliable.
Let's not use the field.
This partially reverts commit a9357c2ce2.
Some kmsg sent before sleep may be received by systemd-journald after
sleep. In that case, map_clock_usec() does not provide correct
timestamp.
So, we cannot provide reliable _SOURCE_MONOTONIC_TIMESTAMP.
Then, fix the monotonic timestamp.
The _SOURCE_MONOTONIC_TIMESTAMP field is already used in other projects.
Hence, we cannot remove the field. But, let's store the correct value.
The existence of the new _SOURCE_BOOTTIME_TIMESTAMP field can indicate
that the monotonic timestamp field is reliable or not.
The timestamp is not in CLOCK_MONOTONIC, but CLOCK_BOOTTIME,
while header monotonic timestamp is in CLOCK_MONOTONIC. Hence, we cannot
adjust timestamp by comparing with header monotonic timestamp and
_SOURCE_MONOTONIC_TIMESTAMP field.
Fixes a regression caused by affde1d7e7.
Fixes#33293.
- add missing assertions,
- use GREEDY_REALLOC() at one more place,
- etc.
Before:
```
$ sudo time valgrind --leak-check=full ./systemd-hwdb update
==112572== Memcheck, a memory error detector
==112572== Copyright (C) 2002-2024, and GNU GPL'd, by Julian Seward et al.
==112572== Using Valgrind-3.23.0 and LibVEX; rerun with -h for copyright info
==112572== Command: ./systemd-hwdb update
==112572==
==112572==
==112572== HEAP SUMMARY:
==112572== in use at exit: 0 bytes in 0 blocks
==112572== total heap usage: 1,320,113 allocs, 1,320,113 frees, 70,614,501 bytes allocated
==112572==
==112572== All heap blocks were freed -- no leaks are possible
==112572==
==112572== For lists of detected and suppressed errors, rerun with: -s
==112572== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
21.94user 0.19system 0:22.23elapsed 99%CPU (0avgtext+0avgdata 229876maxresident)k
0inputs+25264outputs (0major+57275minor)pagefaults 0swaps
```
After:
```
$ sudo time valgrind --leak-check=full ./systemd-hwdb update
[sudo] password for watanabe:
==114732== Memcheck, a memory error detector
==114732== Copyright (C) 2002-2024, and GNU GPL'd, by Julian Seward et al.
==114732== Using Valgrind-3.23.0 and LibVEX; rerun with -h for copyright info
==114732== Command: ./systemd-hwdb update
==114732==
==114732==
==114732== HEAP SUMMARY:
==114732== in use at exit: 0 bytes in 0 blocks
==114732== total heap usage: 1,276,406 allocs, 1,276,406 frees, 68,500,491 bytes allocated
==114732==
==114732== All heap blocks were freed -- no leaks are possible
==114732==
==114732== For lists of detected and suppressed errors, rerun with: -s
==114732== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 0 from 0)
21.91user 0.24system 0:22.26elapsed 99%CPU (0avgtext+0avgdata 233584maxresident)k
0inputs+25168outputs (0major+58237minor)pagefaults 0swaps
```
I do not think this is necessary, but all other places in
libsystemd-network we clear buffer before receive. Without this,
Coverity warns about use-of-uninitialized-values.
Let's silence Coverity.
Closes CID#1469721.
Previously, we did not set lifetime for redirect route, and redirect
routes were removed only when received a RA from the target address.
Thus, routes that redirect on-link addresses were never removed.
RFCs mention nothing about the lifetime of redirection. But the previous
implementation does not pass the IPv6 Core Conformance Tests.
This makes
- remember all received RAs and manage them by the sender address
(previously, remembered only one with the highest preference),
- then use the router lifetime as one for redirect route,
- remove redirect route also when the router corresponds to the sender
address is dropped (previously, considered only target address).
Note, even if we recieve a new RA, we do not update existing redirect
routes. The lifetime of the redirect route is updated only when a new
Redirect message is received.
Closes#32527.
An existing route with lifetime can be always updated with a new
finite lifetime. As the comment in the code says, we cannot disable the
lifetime. So, the condition must be '==', rather than '!='.
Fixes#33210.
It has been mentioned in IPv4Forwarding= and IPv6Forwarding=,
but let's also explain in the settings who imply these settings.
Follow-up for 3976c43092 and
485f5148b3.
PID1 may invoke a helper process, e.g. sd-close, during the test.
So, processes other than networkd and PID1 may be shown.
```
testtap99: tap pi multi_queue vnet_hdr persist filter
Attached to processes:systemd-network(2852)(sd-close)(2863)systemd(1)
```
Closes#33055.
"journalctl -u foo.service" may not work as expected, especially entries
for _TRANSPORT=stdout, for short-living services or when the service manager
generates debugging logs. Instead, SYSLOG_IDENTIFIER= should be reliable for
stdout. Let's use it.
An example case:
```
__CURSOR=s=06278e3bf011458e973c81d370a8f7a5;i=1e4dc;b=1b0258a5c78341609bf462c72d4541c3;m=308de65;t=6194c3895a13f;x=50c7e9af5b8cfc37
__REALTIME_TIMESTAMP=1716665017803071
__MONOTONIC_TIMESTAMP=50912869
_BOOT_ID=1b0258a5c78341609bf462c72d4541c3
SYSLOG_FACILITY=3
_UID=0
_GID=0
_MACHINE_ID=d3490e076ab24968bfa19a6aab26beb3
_HOSTNAME=H
_RUNTIME_SCOPE=system
_TRANSPORT=stdout
PRIORITY=6
_PID=2668
_STREAM_ID=3f9b8855636041988d003a9c63379b8a
SYSLOG_IDENTIFIER=echo
MESSAGE=foo
```
As you can see, there is no unit identifier.
Yu Watanabe