Date: Sun, 25 Apr 1999 00:40:26 +0900 From: "Daniel C. Sobral" <dcs@newsguy.com> To: Andrzej Bialecki <abial@webgiro.com> Cc: Stephane Legrand <Stephane.Legrand@wanadoo.fr>, Luigi Rizzo <luigi@labinfo.iet.unipi.it>, current@FreeBSD.ORG Subject: Re: does login.conf limitations work ? Message-ID: <3721E5EA.672014EB@newsguy.com> References: <Pine.BSF.4.05.9904241637400.25684-100000@freja.webgiro.com>
index | next in thread | previous in thread | raw e-mail
Andrzej Bialecki wrote: > > On Sat, 24 Apr 1999, Stephane Legrand wrote: > > > > That's also my impression. I glipmsed the whole source tree and I couldn't > > > find any place where the limits are enforced. BTW. what entity should > > > enforce login time limits? Kernel? Some user-space daemon? > > > > To report a login.conf success, i've used on a 2.2.8 system the > > "cputime" ressource limit. I set it to zero and that worked very > > well. So may be only some limits are implemented ? > > You're right, this part works. However, I was talking about login time > (which is how long can user be logged on to the system), which isn't > checked anywhere. It isn't. Login.conf, just like, now, pam, is one of those cases of a wonderful clean, orthogonal solution which is only partially implemented. I personally think login time should be enforced through tty revocation. The main problem is that no solution exists which will do the right thing regarding processes to kill for all processes. -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org "Well, Windows works, using a loose definition of 'works'..." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the messagehelp
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3721E5EA.672014EB>