[DTrace-devel] [PATCH] Revert "dt_pid: pid grabs should be shortlived"

[Sam James]

Nick Alcock via DTrace-devel <dtrace-devel at oss.oracle.com> writes:

> On 22 May 2025, eugene loh verbalised:
>
>> From: Eugene Loh <eugene.loh at oracle.com>
>>
>> Commit 4aec5c9931eb ("dt_pid: pid grabs should be shortlived") made
>> grabs short-lived.  No test cases show the utility of this patch.
>> Meanwhile, the patch causes roughly a dozen tests to fail on OL9
>> when the test suite is started remotely using ssh, closing ttys.  The
>
> I think this is a consequence of another bug (Psystem_daemon() was
> failing to detect that systemd was running correctly on systemd v2

(Do you mean cgroup v2?)

> systems, and we were falling back to the old approach, which is indeed
> going to conclude that processes in system groups without controlling
> terminals or associated TTYs are system daemons).
>
> I have a fix for this detection failure under test now and will post it
> shortly.