Re: pre-v2.34.0-rc0 regressions: t7900-maintenance.sh broken due to 'systemd-analyze' (was: What's cooking in git.git (Oct 2021, #06; Mon, 25))

From: Ævar Arnfjörð Bjarmason <avarab@gmail.com>
Date: Oct 26, 2021, 15:34 UTC
Message-ID: <211026.864k93zteh.gmgdl@evledraar.gmail.com>
In-Reply-To: <211026.8635oo11jk.gmgdl@evledraar.gmail.com>

A On Tue, Oct 26 2021, Ævar Arnfjörð Bjarmason wrote:

Show 39 quoted lines

> On Mon, Oct 25 2021, Junio C Hamano wrote:
>
> [Spinning off some "before rc0" threads]
>
>> The fifteenth batch of topics are in 'master'.  I expect that this
>> is more-or-less what we can expect in the -rc0, unless there is a
>> hotfix to what's already merged.
>
> I reported a breakage with $subject in [1], which I see Lénaïc tried to
> fix in [2], but which I managed to miss. Sorry about that.
>
> On that machine (gcc135, on the GCC farm) it's still breaking, relevant
> trace output:
>     
>     ++ printf '%s\n' 'prerequisite SYSTEMD_ANALYZE ok'
>     prerequisite SYSTEMD_ANALYZE ok
>     ++ return 0
>     ++ test_set_prereq SYSTEMD_ANALYZE
>     ++ test -n ''
>     ++ case "$1" in
>     ++ satisfied_prereq=' POSIXPERM BSLASHPSPEC EXECKEEPSPID REFFILES COLUMNS_CAN_BE_1 PERL PTHREADS PYTHON GETTEXT XMLLINT SYSTEMD_ANALYZE '
>     ++ lazily_tested_prereq='EXPENSIVE XMLLINT SYSTEMD_ANALYZE '
>     ++ total_prereq=1
>     ++ case "$satisfied_prereq" in
>     ++ satisfied_this_prereq=t
>     ++ case "$satisfied_this_prereq,$negative_prereq" in
>     ++ ok_prereq=1
>     ++ test 1 = 1
>     ++ systemd-analyze verify systemd/user/git-maintenance@.service
>     Failed to open /dev/tty0: Permission denied
>     Failed to load systemd/user/git-maintenance@.service: Invalid argument
>     error: last command exited with $?=1
>     not ok 34 - start and stop Linux/systemd maintenance
>
> I haven't looked into it, presumably easy-ish to fix, just moving on now
> and checking for other regressions...
>
> 1. https://lore.kernel.org/git/874ka618n4.fsf@evledraar.gmail.com/#t
> 2. https://lore.kernel.org/git/20210927213016.21714-2-lenaic@lhuard.fr/

I looked into this a little bit, the immediate problem is that the prereq testing is different from the command we actually run, and I can run one but not the other on that box for some reason. I.e.

    [avar@gcc135 t]$ systemd-analyze verify /lib/systemd/system/basic.target; echo $?
    Failed to open /dev/tty0: Permission denied
    0
    
    [avar@gcc135 t]$ systemd-analyze verify systemd/user/git-maintenance@.service; echo $?
    Failed to open /dev/tty0: Permission denied
    Failed to load systemd/user/git-maintenance@.service: Invalid argument
    1

So "fixing" that is easy, just have the prereq test that thing in particular, and why does one thing have a /lib/ prefix, but not the other?

But presumably this points to a bigger problem. I.e. we just did a "git maintenance start" a few lines earlier.

If I could start something via systemd then presumably it's either up, or our "start" is buggy and we didn't actually start something, or are using the wrong (non-portable?) invocation to check the running status of the thing we just started?

Also aside from that shouldn't this be:

    test_when_finished "systemd-something stop-it some-service" &&
    systemd-something start-it some-service &&
    [...]

Or are runaway services handled somehow by systemd magic (tied to the PID of the test run?).

So in trying to fix it myself I ran into the boundaries of the little systemd knowledge I've got.