From owner-freebsd-questions Sat Oct 4 00:54:38 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA04613 for questions-outgoing; Sat, 4 Oct 1997 00:54:38 -0700 (PDT) Received: from buffnet4.buffnet.net (buffnet4.buffnet.net [205.246.19.13]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA04608 for ; Sat, 4 Oct 1997 00:54:34 -0700 (PDT) Received: from buffnet7.buffnet.net (shovey@buffnet7.buffnet.net [205.246.19.28]) by buffnet4.buffnet.net (8.7.5/8.7.3) with SMTP id DAA10890; Sat, 4 Oct 1997 03:40:52 -0400 (EDT) Date: Sat, 4 Oct 1997 03:41:16 -0400 (EDT) From: Steve Hovey To: Greg Lehey cc: Dave Hummel , questions@FreeBSD.ORG Subject: Re: slow telnet In-Reply-To: <19971004100251.36524@lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Oh get over yourself! I had pre 2.1Rs that worked fine - put in 2.1R and rlogin puked unless I put stty 9600 in peoples startup files. (actually stty 0 works best but anything less that 9600 makes pine and other jobbers change their visuals) I looked and yes - the rlogin code had CHANGED between versions. You are right - I cant sit here and find the exact hole - or else I woudl have posted the patch - but Im not a complete moron because I cant post a patch. In short - if i feel Im being pissed on, it smells like Im being pissed on, then I probably am even if I cant identify who owned the manly bits originating the water problem. On Sat, 4 Oct 1997, Greg Lehey wrote: > On Fri, Oct 03, 1997 at 08:59:55AM -0400, Steve Hovey wrote: > > On Fri, 3 Oct 1997, Greg Lehey wrote: > > > >> On Thu, Oct 02, 1997 at 02:15:14PM -0400, Steve Hovey wrote: > >>> > >>> I notices here starting with 2.1R that an rlogin would be slow in direct > >>> inverse proportion to the stty baud rate setting. > >>> > >>> So at a $ prompt type stty 9600 and see if it suddenly speeds up. > >> > >> No, this is incorrect. It has no effect on an IP connection. > > > > It does for rlogin - and is most noticible on slower modems. > > How do you measure this? > > This can't work. IP just doesn't work that way. Of course, it's just > barely possible that you have uncovered some bug, but I just couldn't > imagine where. Anyway, just to check, I've set up an rlogin to > FreeBSD.org. I then timed running a 'ps aux'. Here are the results: > > === grog@hub (/dev/ttypk) ~ 2 -> stty -a > speed 9600 baud; 70 rows; 120 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 -dsrflow > -dtrflow -mdmbuf > cchars: discard = ^O; dsusp = ^Y; eof = ^D; eol = ; > eol2 = ; erase = ^H; intr = ^C; kill = ^U; lnext = ^V; > min = 1; quit = ^\; reprint = ^R; start = ^Q; status = ; > stop = ^S; susp = ^Z; time = 0; werase = ^W; > === grog@hub (/dev/ttypk) ~ 3 -> time ps aux > ... > root 3 0.0 0.0 0 12 ?? DL 7Jun97 20:01.45 (vmdaemon) > 1.92 real 0.10 user 0.29 sys > === grog@hub (/dev/ttypk) ~ 6 -> stty speed 50 > 9600 > === grog@hub (/dev/ttypk) ~ 7 -> stty -a > speed 50 baud; 70 rows; 120 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 -dsrflow > -dtrflow -mdmbuf > cchars: discard = ^O; dsusp = ^Y; eof = ^D; eol = ; > eol2 = ; erase = ^H; intr = ^C; kill = ^U; lnext = ^V; > min = 1; quit = ^\; reprint = ^R; start = ^Q; status = ; > stop = ^S; susp = ^Z; time = 0; werase = ^W; > root 2 0.0 0.0 0 12 ?? DL 7Jun97 87:41.77 (pagedaemon) > === grog@hub (/dev/ttypk) ~ 8 -> time ps aux > ... > 0.46 real 0.07 user 0.30 sys > === grog@hub (/dev/ttypk) ~ 9 -> > > The difference in time is coincidental; the net's a bit congested > today. The fact is, stty is talking to a pty, which doesn't transmit > data serially, so there is nowhere for the bit rate to be applied. In > any case, the pty runs only locally. You might just as well set the > bit rate on a console window or an xterm. > > In general, people: if you have a suspicion that a problem may be due > to something you have discovered, and you're not deep gurus in the > area, check the details before you broadcast it to the world. It can > save embarrassment. > > Greg >