Date: Sat, 20 Sep 1997 14:07:10 From: David Goddard <goddard@acm.org> To: Greg Lehey <grog@lemis.com> Cc: questions@freebsd.org Subject: Re: ppp -auto problems and boot-time error messages Message-ID: <3.0.1.16.19970920140710.22dfc0a6@sg1.cc.ic.ac.uk> In-Reply-To: <19970920085241.59762@lemis.com> References: <3.0.1.16.19970919172152.2eb72cb6@sg1.cc.ic.ac.uk> <3.0.1.16.19970919172152.2eb72cb6@sg1.cc.ic.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
At 08:52 20/09/97 +0930, Greg Lehey wrote: >This means you tried to add a default route (probably via >/etc/ppp/ppp.linkup), but one was already set. > >Make sure that /etc/ppp/ppp.linkup contains something like: > >MYADDR: > delete ALL > add 0 0 HISADDR > >I would guess that the 'delete ALL' is missing. Nope - my ppp.linkup was fine, but I've now removed an extra line in ppp.conf: add 0 0 155.198.8.0 This has cleared up the messages, but I'm still having the other problem... [...] >If this is still a problem, I'd need to know more details. Hmmm.. not sure what I can add really. I first noticed it when overnight cron jobs failed. The logs showed the link coming up when the job called it and then timing out after the set inactivity time, while the jobs timed out. Interactive processes also fail. For example, telnet just hangs until I kill it and then when I immediately retype the command, it connects immediately. If I leave it, it times out. This also happens with slrnpull, ftp, ping and anything else I've tried. If it helps, below is a tcpdump record from when I tried to telnet to the host sg1 (which is in my hosts file), with the PPP link originally closed. 155.198.8.0 is the IP address that I arbitarily assigned in ppp.conf and sysconfig - I am assigned a dynamic IP. Output is as follows (wrapped): dmg% tcpdump tcpdump: listening on tun0 01:34:32.241279 155.198.8.0.1024 > sg1.telnet: S 36604141:36604141(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 390 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] 01:34:34.750373 155.198.8.0.1024 > sg1.telnet: S 36604141:36604141(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 395 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] 01:34:40.750298 155.198.8.0.1024 > sg1.telnet: S 36604141:36604141(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 407 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] 01:34:52.750423 155.198.8.0.1024 > sg1.telnet: S 36604141:36604141(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 431 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] 01:35:16.750861 155.198.8.0.1024 > sg1.telnet: S 36604141:36604141(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 479 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] [After this point, the telnet operation times out] [Then, with PPP now up, I successfully repeat the telnet after the first instance has timed out...] 01:36:04.685159 dialup-1-7.net.ic.ac.uk.blackjack > sg1.telnet: S 54414248:54414248(0) win 16384 <mss 1460,nop,wscale 0,nop,nop,timestamp 575 0,nop,nop,ccnew[|tcp]> (DF) [tos 0x10] 01:36:04.850901 sg1.telnet > dialup-1-7.net.ic.ac.uk.blackjack: S 1410624000:1410624000(0) ack 54414249 win 60816 <mss 1460,nop,wscale 0,nop,nop,timestamp 1450969 575> 01:36:04.851519 dialup-1-7.net.ic.ac.uk.blackjack > sg1.telnet: . ack 1 win 17376 <nop,nop,timestamp 576 1450969> (DF) [tos 0x10] 01:36:04.858081 dialup-1-7.net.ic.ac.uk.blackjack > sg1.telnet: P 1:25(24) ack 1 win 17376 <nop,nop,timestamp 576 1450969> (DF) [tos 0x10] 01:36:05.210880 sg1.telnet > dialup-1-7.net.ic.ac.uk.blackjack: . ack 25 win 60792 <nop,nop,timestamp 1450969 576> 01:36:05.485610 dialup-1-7.net.ic.ac.uk.1024 > monolith.cc.ic.ac.uk.domain: 24371+ (42) 01:36:05.771092 monolith.cc.ic.ac.uk.domain > dialup-1-7.net.ic.ac.uk.1024: 24371* 1/5/5 (289) 01:36:05.850945 sg1.telnet > dialup-1-7.net.ic.ac.uk.blackjack: P 1:13(12) ack 25 win 60792 <nop,nop,timestamp 1450971 576> 01:36:05.852759 dialup-1-7.net.ic.ac.uk.blackjack > sg1.telnet: P 25:31(6) ack 13 win 17376 <nop,nop,timestamp 578 1450971> (DF) [tos 0x10] etc. >> I've also rebuilt my kernel since the ppp problem started, but before I >> noticed the boot messages, if this is relevant. > >Probably not. This ppp isn't in the kernel. Just being paranoid :-) >> As an aside, last Saturday I ordered Greg Lehey's FreeBSD book from >> Walnut Creek and it arrived here in the UK yesterday. Needless to say I >> was impressed by the turnaround time. Shame it doesn't have anything on >> networking though :-( > >The next edition will. It'll even contain a reference to exactly this >problem :-) Hopefully I'll have it sorted by them :-) Thanks for your help, Dave
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3.0.1.16.19970920140710.22dfc0a6>