On Tue, 19 Nov 1996, Erik E. Fair wrote:

> You might find the "shlock" program in the original NNTP distribution (last
> version was 1.5) useful in this context...
> I will happily contribute it to NetBSD, if there's interest.

I'd be quite happy to see this in the NetBSD source tree; I've used
it for various things beyond news on my machines. What does core
think?

> One thing: it
> depends upon being able to send signal 0 (null signal), and getting back
> EPERM (permission denied - process exists, but it's not yours), ESRCH (no
> such process), or success (of course, on a null signal, nothing happens).
> HDB UUCP depends on this mechanism to validate locks, too.

This works fine. That's how my routines validate locks.

cjs

Curt Sampson    cjs@portal.ca		Info at http://www.portal.ca/
Internet Portal Services, Inc.	
Vancouver, BC   (604) 257-9400		De gustibus, aut bene aut nihil.