Daan De Meyer [Tue, 15 Feb 2022 11:50:31 +0000 (11:50 +0000)]
test: Stop using TextTestRunner in systemd-networkd-tests.py
If we assign our own test runner, passing arguments stops working
as unittest won't instantiate its own test runner after it parses
the arguments from sys.argv.
Consequence is that the tests will write to stderr now instead of
stdout since it doesn't seem possible to configure the stream that
unittest.main() will instantiate its test runner with so it'll
default to sys.stderr.
Frantisek Sumsal [Mon, 14 Feb 2022 19:19:51 +0000 (20:19 +0100)]
test: set `ReadWritePaths=$BUILD_DIR` when running w/ coverage
otherwise units using `DynamicUser=yes` won't be able to write the
coverage stats (currently affecting TEST-20-MAINPIDGAMES).
`DynamicUser=yes` implies `ProtectSystem=strict` and
`ProtectHome=read-only` and can't be overridden hence we need to
utilize `ReadWritePaths=` to work around that.
When checking if we look at the root directory we actually need to
compare both st_dev *and* st_ino. The existing check only checked the
latter. Fix that.
boot: suppress XBOOTLDR if same device as ESP when enumerating entries
On my local system I linked up the ESP and XBOOTLDR partitions, and
ended up with duplicate entries being listed. Try hard to detect that
and only enumerate entries in the ESP if it turns out that both dirs
have the same dev_t.
This should detect both bind mounted and symlinked cases and should make
our list output less confusing.
bootspec: make sure all return values are initialized on return of find_esp_and_warn()
THis makes sure that find_esp_and_warn() + find_xbootldr_and_warn()
follow our usual coding style that on success all return values are
initialized. We got that right in most successful codepaths out of these
functions, but missed the one where the paths are manually overwritten
via env vars.
bootctl: show more information about boot entry state in list
Let's improve display of boot entries and show what type they have (i.e.
boot loader spec type 1, or type 2, or auto-discovered or reported by
boot loader), and in particular mark entries the boot loader discovered
but we can't find (i.e. that likely vanished, or possibly couldn't be
found due to a misconfiguration) and that the boot loader didn't find
but we see (which are new, or possibly also the result of
misconfiguraiton).
This is supposed to be a replacement for #22161, but instead of hiding
vanished entries, highlights them, which I think is more appropriate for
a low-level tool such bootctl.
Quite often we compare uuids/id128 formatted as strings with specific
values. So far we usually used streq() for that. let's add a new
explicit helper for this in id128_equal_string() that compares a string
with an sd_id128_t and is more robust than a simple string comparison.
Moreover, we can mroe easily reuse the various defines we have for
specific UUIDs, for example those from gpt.h.
We expose various other forms of UUID helpers already, i.e.
SD_ID128_UUID_FORMAT_STR and SD_ID128_MAKE_UUID_STR(), and we parse
UUIDs, hence add a high-level helper for formatting UUIDs too.
This doesn't add any new code, it just moves some helpers
id128-util.[ch] → sd-id128.[ch], to make them public.
gpt-auto: properly handle case where we can't determine devno of /usr/ fs
get_block_device_harder() returns == 0 if the fs is valid, but it is not
backed by a single devno. (As opposed to returning > 0 if the devno is
valid). Let's catch this case and log a clear message, and don't bother
open the device in that case.
This is mostly cosmetical, as either way, systemd-gpt-auto-generator
doesn't work in scenarios like that.
sd-device: refuse opening device mit major/minor of zero early
device nodes with 0 dev_t are not real (and for that reason such inodes
are used as whiteouts in overlayfs, for example), hence refuse them
early. It seems wrong going to sysfs for something we know can't exist
anyway.
Frantisek Sumsal [Mon, 14 Feb 2022 11:20:02 +0000 (12:20 +0100)]
test: check for (possibly) missing test coverage
If the test logs contain lines like:
```
...systemd-resolved[735885]: profiling:/systemd-meson-build/src/shared/libsystemd-shared-250.a.p/base-filesystem.c.gcda:Cannot open
```
it means we're possibly missing some coverage since gcov can't write the stats,
usually due to the sandbox being too restrictive (e.g. ProtectSystem=yes,
ProtectHome=yes) or the $BUILD_DIR being inaccessible to non-root users.
Yu Watanabe [Thu, 10 Feb 2022 07:14:21 +0000 (16:14 +0900)]
test-dhcp6-client: cleanups
This does many cleanups. Most prominently,
- check the sents packet more strictly,
- make the packets used in the test more readable,
- unify callbacks.
Yu Watanabe [Mon, 7 Feb 2022 15:30:56 +0000 (00:30 +0900)]
fuzz-dhcp6-client: test multiple states
Then, now all three paths client_process_information(),
client_process_reply(), and client_process_advertise_or_rapid_commit_reply()
in client_receive_message() are fuzzed.
Yu Watanabe [Thu, 10 Feb 2022 12:44:28 +0000 (21:44 +0900)]
sd-dhcp6-client: append extra options before elapsed time option
The value of elapsed time option is determined in runtime and not
deterministic. It is hard to test the sent packets if it is located
in the intermidiate of the packet.
Yu Watanabe [Mon, 7 Feb 2022 14:36:49 +0000 (23:36 +0900)]
network: do not restart DHCPv6 client when it is already running in managed mode
Otherwise, if a network has two router and one provides M flag and the
other provides O flag, then the DHCPv6 client switches the running mode
repeatedly. That makes the IPv6 network configuration of the host
unstable.
Yu Watanabe [Sun, 6 Feb 2022 17:48:03 +0000 (02:48 +0900)]
sd-dhcp6-client: drop unnecessary event_source_disable()
The event source will be soon re-enabled in the same function.
The function client_timeout_resend() may return earlier without
re-enabling the timer souce. However,
- the timer event source is one shot by default. Hence, it is not
necessary to disable in the callback function,
- when it returns early, then client_set_state() or client_stop() is
called before return, and they re-ernable or disable the timer.
Yu Watanabe [Sun, 6 Feb 2022 17:36:55 +0000 (02:36 +0900)]
sd-dhcp6-client: set lease expiration timer in client_enter_bound_state()
The timer must be active until the client get re-enter bound state,
and the timeout must be determined by the lease acquired when entering
the bound state.
Yu Watanabe [Sun, 6 Feb 2022 14:51:43 +0000 (23:51 +0900)]
sd-dhcp6-client: fix lifetime handling
This fixes the following issues:
- if T1 is finite and T2 is infinite, then the timer for T1 was not
enabled,
- after T1 and T2 are randomized, T1 may be longer than T2.
projects / thirdparty / systemd.git / log
