From owner-freebsd-hackers Mon Jun 1 03:15:38 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id DAA08393 for freebsd-hackers-outgoing; Mon, 1 Jun 1998 03:15:38 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from spinner.netplex.com.au (spinner.netplex.com.au [202.12.86.3]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id DAA08375; Mon, 1 Jun 1998 03:15:15 -0700 (PDT) (envelope-from peter@netplex.com.au) Received: from spinner.netplex.com.au (localhost [127.0.0.1]) by spinner.netplex.com.au (8.8.8/8.8.8/Spinner) with ESMTP id SAA08208; Mon, 1 Jun 1998 18:14:46 +0800 (WST) (envelope-from peter@spinner.netplex.com.au) Message-Id: <199806011014.SAA08208@spinner.netplex.com.au> X-Mailer: exmh version 2.0.2 2/24/98 To: Atsushi Murai cc: hackers@FreeBSD.ORG Subject: Re: cvs commit: src/lib/libutil libutil.h logwtmp.c src/usr.bin/login login.c In-reply-to: Your message of "Mon, 01 Jun 1998 01:47:05 MST." <199806010847.BAA12337@freefall.freebsd.org> Date: Mon, 01 Jun 1998 18:14:46 +0800 From: Peter Wemm Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Atsushi Murai wrote: > amurai 1998/06/01 01:47:05 PDT > Trim a domain part for wtmp as same as showed by "netstat -r". > Here is a some example for avoiding a confusion. Incidently, I think we should double the UT_HOSTNAME length as well. That means we end up with a 64 byte utmp record instead of the original 32 byte record in 2.x. BSDI also doubled the UT_HOSTNAME field, if we followed then we'd have compatable utmp/wtmp files again, and BSD/OS binaries could read ours too. I'm not sure I want to go through _another_ utmp record size change again though..... :-( I might also add that I think utmp/wtmp is a horrible hack. It would be great if it could be redesigned with a real API for reading/writing it. It would be nice to have a SYSV/Linux/Xopen/Unix98 compatable API too. A utmp.conf file might also be nice if it could configure the record sizes. Alternatively, perhaps use a DB utmp.db file with variable record sizes keyed on hashed tty name. wtmp.db would need something more space efficient than a hash method file and would need to maintain sequence, perhaps the btree DB method? As long as there was a get/put{u,w}tmp style API and logwtmp(), then that should cover most incompatability problems. Since libutil is shared, we could update that library and it would automatically update many of the utmp-using binaries to use the new format. Sigh, oh the things that could be done with infinite spare time... :-) Cheers, -Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message