Date: Mon, 03 Dec 2001 21:48:19 -0500 From: James Housley <jim@thehousleys.net> To: Greg Lane <gregory.lane@anu.edu.au> Cc: freebsd-hackers@freebsd.org, freebsd-questions@freebsd.org Subject: Re: Jul 30 changes to ppp break my Telstra ADSL PPPoE connection Message-ID: <3C0C3973.3DEBE8F7@Thehousleys.net> References: <20011204105316.A25712@nucl03.anu.edu.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Greg Lane wrote: > > G'day all, > > I have a problem with PPPoE over Telstra ADSL (Australia). A full > description of the problem can be found in freebsd-questions > (see 20011201011107.A9404@nucl03.anu.edu.au dated Nov 30) > with the subject: > > ADSL PPPoE (Telstra Aust.) began to hang between 4.3- and 4.4-RELEASE > > The short of it is that at some point during my cvsup history certain > types of traffic would hang, specifically most web traffic and certain > types of ssh (e.g. ls -l of a long directory). This was reminiscent of > the MTU/MRU problems at the beginning of the year. I backdated to > 4.3-RELEASE which worked and I forgot about it until recently when I > decided to change my gateway machine. > > I have now chased it up in earnest by going back through > various dated "cvsup's" and making the world. On > July 30th there were some changes to src/usr.sbin/ppp that break ppp > and cause this problem. Strangely enough the changes were related to > MRU's :-> !! > > I can reliably make my DSL connection work on three different machines > using four different network cards (ed, de, vr, rl) with source dated > Jul 30 12:01am, and reliably break it with anything after > Jul 31 12:01am. > > The changes made to ppp on that day were to: > > src/usr.sbin/ppp/command.c > src/usr.sbin/ppp/lpc.c > src/usr.sbin/ppp/nat_cmd.c > src/usr.sbin/ppp/nat_cmd.h > src/usr.sbin/ppp/ppp.8 > > The most likely problem is lpc.c because 1) it looks the most > complicated(!!) and 2) it is supposed to: > > MFC: Handle peer REQ/NAKs of >1500 byte MRUs when we have no preference > > The other diffs were small and were to implement two issues: > > MFC: If the peer REJects our MRU REQ, stop REQing it -- *EVEN* if we're > doing PPPoE and the default MRU is therefore too big. > > MFC: Actually add the ``nat proto'' command > > I'm afraid my lack of knowledge stops me going on from here. My > ppp.conf is below and I start ppp with "ppp -ddial -nat -quiet bigpond" > Can anyone see the problem and suggest further detective work I can > perform or even better, a fix or workaround? > > Cheers, > Greg > > P.S. I have cross-posted to questions since the thread started there > and hackers since I believe I have narrowed down the problem and it is > technical. I hope this is OK. Please confine your reply to whatever > list you feel is appropriate, I am subscribed to both. > > --------------------------------------------------- > default: > #Only enable logging for troubleshooting > #set log CBCP CCP Chat Connect Command IPCP tun Phase Warning Debug LCP sync > > set device PPPoE:rl0:bigpond > set speed sync > set mru 1454 > set mtu 1454 I had a similar problem, but haven't been able to fully track it down. I did file a PR http://www.freebsd.org/cgi/query-pr.cgi?pr=32040 . A LCP log of both versions was requested. Also if you read the new man page there is "set mtu max x" and "set mru max x" options, the I might have mistyped when I tried. Jim -- /"\ ASCII Ribbon Campaign . \ / - NO HTML/RTF in e-mail . X - NO Word docs in e-mail . / \ ----------------------------------------------------------------- jeh@FreeBSD.org http://www.FreeBSD.org The Power to Serve jim@TheHousleys.Net http://www.TheHousleys.net jhousley@SimTel.Net http://www.SimTel.Net --------------------------------------------------------------------- If it happens once, it's a bug. If it happens twice, it's a feature. If it happens more than twice, it's windows. -- Luiz de Barros To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C0C3973.3DEBE8F7>