[26507] in Source-Commits
Re: /svn/athena r25589 - trunk/debathena/config/reactivate/debian
daemon@ATHENA.MIT.EDU (Benjamin Kaduk)
Sun Jul 1 23:38:00 2012
Date: Sun, 1 Jul 2012 23:37:57 -0400 (EDT)
From: Benjamin Kaduk <kaduk@MIT.EDU>
To: Jonathan Reed <jdreed@MIT.EDU>
cc: source-commits@MIT.EDU
In-Reply-To: <D8724FF8-B180-4B06-A0DC-1D4D150A132F@MIT.EDU>
Message-ID: <alpine.GSO.1.10.1207012336200.18441@multics.mit.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
On Fri, 29 Jun 2012, Jonathan Reed wrote:
>> The kernel programmer in me would want to repeat the same-root checks
>> here, for paranoia's sake.
>>
>> But that may be excessive, here.
>
> We're unlikely to cycle through 32768 pids between the TERM and the
> KILL, but I can add it if you'd prefer. Otherwise, apathy will win.
Apathy is fine ... if someone manages to fire up a forkbomb, I expect the
machine will be rebooting anyway.
And we really shouldn't try to get in the business of working around
obscure kernel bugs (I remember a story when the kernel would reuse a pid
before it should have, due to some race condition or something).
-Ben
