Red Hat Bugzilla – Bug 7784
Updated RH6.1 dials twice for PPP
Last modified: 2008-05-01 11:37:53 EDT
Hello, I also reported bug 6324 which is related to this. I've just installed a fresh RH6.1 with these updates: > ppp-2.3.10-3.i386.rpm > rp3-1.0.1-1.i386.rpm > initscripts-4.70-1.i386.rpm After that I configured the PPP setup for my cable modem, and then it shows the familiar problem of dialling twice before connecting. Disconnecting and reconnecting works OK. This is the output in /var/log/messages, with the IP addresses cleared for security: Dec 12 19:38:54 localhost ifup-ppp: pppd started for ppp0 on /dev/ttyS0 at 115200 Dec 12 19:38:54 localhost modprobe: can't locate module char-major-108 Dec 12 19:38:54 localhost kernel: CSLIP: code copyright 1989 Regents of the University of California Dec 12 19:38:54 localhost kernel: PPP: version 2.3.7 (demand dialling) Dec 12 19:38:54 localhost kernel: PPP line discipline registered. Dec 12 19:38:54 localhost kernel: registered device ppp0 Dec 12 19:38:54 localhost pppd[663]: pppd 2.3.10 started by root, uid 0 Dec 12 19:39:41 localhost pppd[663]: Connect script failed Dec 12 19:39:42 localhost pppd[663]: Exit. Dec 12 19:39:42 localhost ifup-ppp: pppd started for ppp0 on /dev/ttyS0 at 115200 Dec 12 19:39:42 localhost modprobe: can't locate module char-major-108 Dec 12 19:39:42 localhost pppd[677]: pppd 2.3.10 started by root, uid 0 Dec 12 19:39:43 localhost pppd[677]: Serial connection established. Dec 12 19:39:43 localhost pppd[677]: Using interface ppp0 Dec 12 19:39:43 localhost pppd[677]: Connect: ppp0 <--> /dev/ttyS0 Dec 12 19:39:45 localhost kernel: PPP BSD Compression module registered Dec 12 19:39:45 localhost kernel: PPP Deflate Compression module registered Dec 12 19:39:45 localhost pppd[677]: local IP address xxx.xx.xx.xx Dec 12 19:39:45 localhost pppd[677]: remote IP address yyy.yy.yy.yy Dec 12 19:39:45 localhost pppd[677]: primary DNS address yyy.yy.yy.yy Dec 12 19:39:45 localhost pppd[677]: secondary DNS address zzz.zz.zz.zz Thanks for your support, kind regards, Pim Vijfhuizen vijfhuizen@nl.ibm.com
Can you turn on debug mode (add debug to /etc/ppp/options), so we can get a better idea where it's failing?
Here is the requested debug output. I hope this helps. Dec 22 22:59:23 linuxbox ifup-ppp: pppd started for ppp0 on /dev/ttyS0 at 115200 Dec 22 22:59:23 linuxbox modprobe: can't locate module char-major-108 Dec 22 22:59:23 linuxbox kernel: CSLIP: code copyright 1989 Regents of the University of California Dec 22 22:59:23 linuxbox kernel: PPP: version 2.3.7 (demand dialling) Dec 22 22:59:23 linuxbox kernel: PPP line discipline registered. Dec 22 22:59:23 linuxbox kernel: registered device ppp0 Dec 22 22:59:23 linuxbox pppd[561]: pppd 2.3.10 started by root, uid 0 Dec 22 22:59:24 linuxbox chat[574]: send (ATZ^M) Dec 22 22:59:24 linuxbox chat[574]: expect (OK) Dec 22 23:00:09 linuxbox chat[574]: alarm Dec 22 23:00:09 linuxbox chat[574]: Failed Dec 22 23:00:09 linuxbox pppd[561]: Connect script failed Dec 22 23:00:10 linuxbox pppd[561]: Exit. Dec 22 23:00:10 linuxbox modprobe: can't locate module char-major-108 Dec 22 23:00:10 linuxbox ifup-ppp: pppd started for ppp0 on /dev/ttyS0 at 115200 Dec 22 23:00:10 linuxbox pppd[577]: pppd 2.3.10 started by root, uid 0 Dec 22 23:00:11 linuxbox chat[587]: send (ATZ^M) Dec 22 23:00:11 linuxbox chat[587]: expect (OK) Dec 22 23:00:11 linuxbox chat[587]: ^M Dec 22 23:00:11 linuxbox chat[587]: NO CARRIER^M Dec 22 23:00:11 linuxbox chat[587]: ^M Dec 22 23:00:11 linuxbox chat[587]: ATZ^M Dec 22 23:00:11 linuxbox chat[587]: OK Dec 22 23:00:11 linuxbox chat[587]: -- got it Dec 22 23:00:11 linuxbox chat[587]: send (ATDT4^M) Dec 22 23:00:11 linuxbox chat[587]: expect (CONNECT) Dec 22 23:00:11 linuxbox chat[587]: ^M Dec 22 23:00:11 linuxbox chat[587]: ATDT4^M Dec 22 23:00:11 linuxbox chat[587]: CONNECT Dec 22 23:00:11 linuxbox chat[587]: -- got it Dec 22 23:00:11 linuxbox chat[587]: send (^M) Dec 22 23:00:11 linuxbox chat[587]: timeout set to 5 seconds Dec 22 23:00:11 linuxbox chat[587]: expect (~) Dec 22 23:00:11 linuxbox chat[587]: ^M Dec 22 23:00:12 linuxbox chat[587]: ~ Dec 22 23:00:12 linuxbox chat[587]: -- got it Dec 22 23:00:12 linuxbox chat[587]: send (^M) Dec 22 23:00:12 linuxbox pppd[577]: Serial connection established. Dec 22 23:00:12 linuxbox pppd[577]: Using interface ppp0 Dec 22 23:00:12 linuxbox pppd[577]: Connect: ppp0 <--> /dev/ttyS0 Dec 22 23:00:16 linuxbox kernel: PPP BSD Compression module registered Dec 22 23:00:17 linuxbox kernel: PPP Deflate Compression module registered Dec 22 23:00:17 linuxbox pppd[577]: local IP address aaa.aa.aa.aaa Dec 22 23:00:17 linuxbox pppd[577]: remote IP address bbb.bb.bb.bbb Dec 22 23:00:17 linuxbox pppd[577]: primary DNS address ccc.cc.cc.cc Dec 22 23:00:17 linuxbox pppd[577]: secondary DNS address ddd.dd.dd.dd Kind regards, Pim Vijfhuizen vijfhuizen@nl.ibm.com
I've made a workaround by inserting a "TIMEOUT 5" statement in the CHAT script, just before the ATZ modem init string. This way the script doesn't wait 45 seconds before failing, but only 5. The redial starts sooner and succeeds. This eliminates 40 seconds of annoying wait time. Did I do something wrong in reporting? I believe this bug wasn't high on the priority list.
Does the ppp in rawhide/6.2beta work any better? I believe there was *another* case where it would get itself killed.
Closing at customer's request.
This defect has been resolved in a newer release.