Re: Big compat issue with a recent current (zfs + syscall)

From: Xin LI <delphij_at_gmail.com>
Date: Thu, 21 Aug 2025 23:29:08 UTC
On Thu, Aug 21, 2025 at 2:56 PM Kyle Evans <kevans@freebsd.org> wrote:
[...]

> I don't think I'd expect it to help with zfs dataset stuff.
>

There are also some recent ZFS changes, but I don't know (didn't have read
through all changes) if there are something related.


> >> I had wondered the same, but the use of 'segfault' gave me pause; these
> would be SIGSYS rather than SIGSEGV, but that could just be a minor
> terminology dispute.
> >
> > Aug 20 10:35:32 Andromeda kernel: [566445] pid 52166 (auth), jid 50, uid
> 143: exited on signal 6 (no core dump - sugid process denied by ke
> > rn.sugid_coredump)
> > Aug 20 10:35:37 Andromeda kernel: [566450] pid 52172 (auth), jid 50, uid
> 143: exited on signal 6 (no core dump - sugid process denied by ke
> > rn.sugid_coredump)
> > Aug 20 10:35:44 Andromeda kernel: [566457] pid 52179 (auth), jid 50, uid
> 143: exited on signal 6 (no core dump - sugid process denied by ke
> > rn.sugid_coredump)
> > Aug 20 10:35:51 Andromeda kernel: [566463] pid 52185 (auth), jid 50, uid
> 143: exited on signal 6 (no core dump - sugid process denied by ke
> > rn.sugid_coredump)
> > Aug 20 10:35:56 Andromeda kernel: [566469] pid 52193 (auth), jid 50, uid
> 143: exited on signal 6 (no core dump - sugid process denied by ke
> > rn.sugid_coredump)
> >
>
> SIGABRT would seem to imply something like an assertion being tripped,
> which is a bit unusual.  Might need to flip
> kern.sugid_coredump for a minute and see if you can gather some more
> context from a coredump.
>

Well, yes (it's somewhat unusual) and no (I think it's expected behavior
here):  It's not unusual for a program to explicitly abort() when a
setgroups() call failed, though: if I was the programmer who wanted to drop
privileges and failed, showing some error message and abort() as soon as
possible would be a reasonable choice (IMHO) because I wouldn't have a lot
of other remedies, so to me it sounded reasonable here, especially when
Alexander saw some missing system calls earlier.

Cheers,