From owner-freebsd-questions Wed Jan 11 05:13:42 1995 Return-Path: questions-owner Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id FAA28171 for questions-outgoing; Wed, 11 Jan 1995 05:13:42 -0800 Received: from desiree.teleport.com (desiree.teleport.com [192.108.254.11]) by freefall.cdrom.com (8.6.9/8.6.6) with ESMTP id FAA28143; Wed, 11 Jan 1995 05:13:15 -0800 From: bmk@dtr.com Received: (from uucp@localhost) by desiree.teleport.com (8.6.9/8.6.9) with UUCP id FAA14726; Wed, 11 Jan 1995 05:13:11 -0800 Received: (from bmk@localhost) by dtr.com (8.6.9/8.6.9) id BAA10962; Tue, 10 Jan 1995 01:40:13 -0800 Message-Id: <199501100940.BAA10962@dtr.com> Subject: phantom users To: questions@FreeBSD.org, bugs@FreeBSD.org Date: Tue, 10 Jan 1995 01:40:13 -0800 (PST) Reply-To: bmk@dtr.com X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 4920 Sender: questions-owner@FreeBSD.org Precedence: bulk A few months ago, I had a problem under 1.1 and 1.1.5.1 with users showing as logged in on sio ports long after they'd logged out. I observed this under 2.0 shortly after I upgraded. The problem seemed to cure itself, so I forgot about it. Today I noticed it again. I did a 'who': bmk tty00 Jan 10 00:01 uuramjet tty01 Jan 9 22:55 The bmk entry is valid... I am indeed logged in on tty00. uuramjet is a phantom entry. uuramjet is a uucp login, and it's been logged out for almost three hours, as is proved by my uucp log: uucico - - (1995-01-09 22:55:12.92 6054) Incoming call (login uuramjet port tty01) uucico ramjet - (1995-01-09 22:55:13.50 6054) Handshake successful (protocol 'i' sending packet/window 1024/16 receiving 1024/16) uucico ramjet root (1995-01-09 22:55:14.75 6054) Receiving rmail ramjet.dtr.com!roger (1763 bytes) uucico ramjet root (1995-01-09 22:55:15.62 6054) Receiving rmail ramjet.dtr.com!rjc (1318 bytes) uucico ramjet root (1995-01-09 22:55:16.34 6054) Receiving rmail ramjet.dtr.com!roger (1696 bytes) uucico ramjet - (1995-01-09 22:55:17.04 6054) Protocol 'i' packets: sent 8, resent 0, received 14 uucico ramjet - (1995-01-09 22:55:17.24 6054) Call complete (5 seconds 4777 bytes 955 bps) uuxqt ramjet root (1995-01-09 22:55:19.85 9679) Executing X.ramjetC0044 (rmail ramjet.dtr.com!roger) uuxqt ramjet root (1995-01-09 22:55:20.92 9679) Executing X.ramjetC0047 (rmail ramjet.dtr.com!rjc) uuxqt ramjet root (1995-01-09 22:55:21.84 9679) Executing X.ramjetC004a (rmail ramjet.dtr.com!roger) uux ramjet daemon (1995-01-09 23:00:17.76 9703) Queuing rmail rjc@ramjet.dtr.com (D.001N) uux ramjet daemon (1995-01-09 23:00:18.78 9704) Queuing rmail roger@ramjet.dtr.com (D.001Q) uux ramjet daemon (1995-01-09 23:00:19.83 9705) Queuing rmail roger@ramjet.dtr.com (D.001T) Just for kicks, here's the output of last | head -20 (with some notes): bmk tty00 Tue Jan 10 00:01 still logged in # Here's the entry I'm talking about. uuramjet tty01 Mon Jan 9 22:55 still logged in # Here it is again. I was only on for about an hour. bmk tty00 Mon Jan 9 21:10 - 00:01 (02:51) root ttyv1 Mon Jan 9 18:56 - 19:27 (00:30) bmk ttyv1 Mon Jan 9 18:15 - 18:56 (00:40) bmk ttyv0 Mon Jan 9 18:10 - 19:35 (01:24) reboot ~ Mon Jan 9 18:10 # it happened here. It's a long story, but I had to reset the # machine at 18:10. uuramjet had been logged out for a long time. Each of the following five connections were very short ones. uuramjet tty01 Mon Jan 9 14:16 - crash (03:53) uuramjet tty01 Mon Jan 9 13:50 - 14:16 (00:26) uuramjet tty01 Mon Jan 9 13:27 - 13:50 (00:22) uuramjet tty01 Mon Jan 9 12:35 - 13:27 (00:52) uuramjet tty01 Mon Jan 9 12:29 - 12:35 (00:05) reboot ~ Mon Jan 9 12:26 shutdown ~ Mon Jan 9 12:07 bmk ttyp1 k2 Mon Jan 9 11:42 - shutdown (00:24) reboot ~ Mon Jan 9 11:41 shutdown ~ Mon Jan 9 06:20 root ttyv1 Mon Jan 9 05:46 - shutdown (00:33) bmk ttyv1 Mon Jan 9 05:33 - 05:46 (00:12) root ttyv0 Mon Jan 9 05:09 - shutdown (01:11) And for your viewing pleasure, here's the stty settings on the port; although I doubt it's relavent. speed 115200 baud; 0 rows; 0 columns; lflags: -icanon -isig -iexten -echo -echoe -echok -echoke -echonl -echoctl -echoprt -altwerase -noflsh -tostop -flusho -pendin -nokerninfo -extproc iflags: -istrip -icrnl -inlcr -igncr -ixon -ixoff -ixany -imaxbel -ignbrk -brkint -inpck -ignpar -parmrk oflags: -opost -onlcr -oxtabs cflags: cread cs8 -parenb -parodd hupcl -clocal -cstopb crtscts -mdmbuf cchars: discard = ^O; dsusp = ^Y; eof = ^D; eol = ; eol2 = ; erase = ^?; intr = ^C; kill = ^U; lnext = ^V; min = 1; quit = ^\; reprint = ^R; start = ^Q; status = ; stop = ^S; susp = ^Z; time = 0; werase = ^W; I can find no rhyme or reason to this. If this was just my personal system, I wouldn't care. The problem is that I need to account for every minute of login time, ESPECIALLY that done on sio ports. I'd be happy to try and hunt this bug down, even though I'm not a C wizard, but I'm very desparate to get a fix for this. Has anyone else seen this, and is there some kind of fix or workaround? Where should I start looking for this possible bug? It looks to me that wtmp is not always getting updated when the user logs out, but it never seems to happen on ptys. HELP! Free pizza and beer to the person who can help me fix this... but you'll have to visit Portland to collect!