Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 2 Apr 2001 14:29:32 -0700
From:      "Matthew H. North" <ctsmhn@cts.com>
To:        "Jeremiah Gowdy" <jgowdy@home.com>, <freebsd-questions@FreeBSD.ORG>
Subject:   RE: Reproducible kernel panics, 4.2-STABLE, various hardware
Message-ID:  <LAEKJNMMIOPFCDDPFBDEKEBICBAA.ctsmhn@cts.com>
In-Reply-To: <006b01c0bbb7$baed3150$035778d8@sherline.net>

next in thread | previous in thread | raw e-mail | index | archive | help

| That's not true though.  Userland programs running as root can
| cause kernel
| panics and hard lockups.  :)

Interesting.  Is that because root-running programs have direct access to
hardware and BIOS?  Or is this true even if the program only uses system
calls to get anything low-level done?

In any event, the way squid works is it starts as root and establishes
itself as a parent process, then spawns a child that uses seteuid (and
setegid) to run as an unprivileged user.  The root-running parent just
watches its single child and spawns a new one should it die.  That should
make the point moot.

But I was just looking at the squid code and it appears that it has a
function that changes uid back to 0.  Not sure if children are allowed to
use that function or not.  Just another ? ... sigh.

|
| Have you tried CVSUPing to 4.3-RC and seeing if it does it there?

I have not - I'll have to try that next I suppose.

- Matt


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?LAEKJNMMIOPFCDDPFBDEKEBICBAA.ctsmhn>