From owner-freebsd-current Mon Dec 18 21:56:04 1995 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA24766 for current-outgoing; Mon, 18 Dec 1995 21:56:04 -0800 (PST) Received: from localhost (vet.vet.purdue.edu [128.210.79.50]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA24729 for ; Mon, 18 Dec 1995 21:55:58 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by localhost (8.6.12/8.6.12) with SMTP id AAA04015 for ; Tue, 19 Dec 1995 00:55:57 -0500 Message-Id: <199512190555.AAA04015@localhost> X-Authentication-Warning: localhost.vet.purdue.edu: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 To: FreeBSD-current@freebsd.org Subject: Re: ppp From: Benjamin Lewis Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 19 Dec 1995 00:55:54 -0500 Sender: owner-current@freebsd.org Precedence: bulk > On Mon, 18 Dec 1995, Paul Richards wrote: > > > Whatever the cause, dial-on-demand is broken and there's at least two > > of us seeing the problem. > > dial-on-demand, user land ppp, is solid for me. i am running > 2.1R with a sportster 14.4 and asus sp3g on-board uarts. > Just to add another weirdness to the pile, iijppp doesn't do demand dialing for me either, but in a different way than I've seen other people mention. I can use it to do a regular dialin with no problem... well, I did have to do a bit of hacking on the source to get it to work 100%. My login script is extremely long (I have to log in to an annex, log into a computer, bounce back to the annex to set it to do 8 bits properly, bounce back to the computer and start up SLiRP), and it wasn't sending the last few characters, and thus failed to actually start up the PPP process. I hacked the length of the array that contains the script (was a long time ago, don't remember details), recompiled, and all was peachy keen for vanilla usage. I could type "ppp vet" and it would happily start the ppp process and everything was great. However, if I tried "ppp -demand vet" and sent a packet, it would dial up, log in, start the remote ppp, and core dump. At that point I gave up hacking on it, and went with pppd. I figured (still do) that either I broke something when I hacked the array length, or that some other weirdness peculiar to my system is responsible, especially as I didn't see hordes of complaints on the newsgroups and mailing lists. Now, I think that maybe my problem might just be an exaggerated version of other people's difficulties. Perhaps the extra long script is just tickling a bug a little harder or something. Since my C coding skills were just about exhuasted by changing [512] to [1024] or whatever it was, I can't be of any more help than sending copies of my config files to interested parties. Oh, the original ppp binary I had was from the last 2.1-SNAP, as is the rest of my userland stuff, but I went and snagged the 2.1.0-RELEASE iijppp source code for my hacking. -Ben "waiting for a kernel written in TCL so *I* can hack on it :)" Lewis -- Benjamin Lewis (blewis@vet.vet.purdue.edu)