Listmates, I have a server (Box 2) that has a modem that is configured to act like an "ISP" and accept dial-in connections and start a ppp session to provide an internet connection to a user that has dialed in from Box 1 by giving the user the ability to access the internet through the broad-band connection the server has. For years this has just worked on 10.0 and 10.3, now for some reason I can't make it work anymore on 11.0. To help visualize the simple setup here is an attempt at an ascii drawing: |----------------| |---------------| | Client - Box 1 | | Server - Box2 | | openSuSE 10.3 | | openSuSE 11.0 | | | Serial Con. | | | Internal |<----------->| Internal | | Modem | | Modem | |----------------| | | Internet | Broad Band |<----------> | eth0 | |---------------| Box 1 uses wvdial and kinternet to connect to Box 2. There is no problem connecting, but for some reason ppp acts like it has lost its mind and finally terminates the connection. I can dial in with minicom and work on Box 2 without any problem and I can use lynx to access the net, etc.. but for some reason I cannot get ppp to work over the "Serial Con." link shown above. I have tried just about ever combination of option in /etc/ppp/options and /etc/mgetty+sendfax/login, to no avail. A summary of what the kinternet log shows is: SuSE Meta pppd (smpppd-ifcfg), Version 1.59 on lakehouse. Status is: disconnected trying to connect to smpppd connect to smpppd Status is: disconnected Status is: connecting pppd[0]: Plugin passwordfd.so loaded. pppd[0]: --> WvDial: Internet dialer version 1.56 (abuild@balli) pppd[0]: --> Initializing modem. pppd[0]: --> Sending: ATZ pppd[0]: ATZ pppd[0]: OK pppd[0]: --> Sending: AT Q0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 pppd[0]: AT Q0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 pppd[0]: OK pppd[0]: --> Sending: ATM1 pppd[0]: ATM1 pppd[0]: OK pppd[0]: --> Modem initialized. pppd[0]: --> Sending: ATDTxxxxxxx pppd[0]: --> Waiting for carrier. pppd[0]: ATDTxxxxxxx pppd[0]: CONNECT 28800/ARQ/V34/LAPM/V42BIS pppd[0]: --> Carrier detected. Waiting for prompt. pppd[0]: Welcome to openSUSE 11.0 (i586) - Kernel pppd[0]: (l). pppd[0]: nemesis!login: pppd[0]: Welcome to openSUSE 11.0 (i586) - Kernel pppd[0]: (l). pppd[0]: nemesis!login: pppd[0]: --> Looks like a login prompt. pppd[0]: --> Sending: david pppd[0]: david pppd[0]: Password: pppd[0]: --> Looks like a password prompt. pppd[0]: --> Sending: (password) pppd[0]: Last login: Mon Sep 7 17:13:40 CDT 2009 on ttyS1 pppd[0]: You have new mail. pppd[0]: Have a lot of fun... pppd[0]: [1b](K[1b][0;37m17:22[1b][1;34m [1b][1;34mnemesis:~> [1b][0m pppd[0]: --> Don't know what to do! Starting pppd and hoping for the best. pppd[0]: Serial connection established. pppd[0]: using channel 3 pppd[0]: Renamed interface ppp0 to modem0 pppd[0]: Using interface modem0 Status is: connecting pppd[0]: Connect: modem0 <--> /dev/ttyS1 pppd[0]: sent [LCP ConfReq id=0x1 <asyncmap 0x200a0000> <magic 0xaaa92ff3> <pcomp> <accomp>] pppd[0]: rcvd [LCP ConfReq id=0x1 <asyncmap 0x200a0000> <magic 0xaaa92ff3> <pcomp> <accomp>] pppd[0]: sent [LCP ConfNak id=0x1 <magic 0xe117d89f>] <snip more magic number negotiation> pppd[0]: rcvd [LCP ConfReq id=0xa <asyncmap 0x200a0000> <magic 0xfc94a72f> <pcomp> <accomp>] pppd[0]: sent [LCP ConfNak id=0xa <magic 0xea920920>] pppd[0]: rcvd [LCP ConfNak id=0xa <magic 0xea920920>] pppd[0]: Serial line is looped back. pppd[0]: sent [LCP TermReq id=0xb "Loopback detected"] pppd[0]: sent [LCP TermReq id=0xc "Loopback detected"] pppd[0]: rcvd [LCP TermReq id=0xc "Loopback detected"] pppd[0]: sent [LCP TermAck id=0xc] pppd[0]: Connection terminated. pppd[0]: Hangup (SIGHUP) pppd[0]: Modem hangup Status is: disconnected pppd[0] died: A modem hung up the phone (exit code 16) Which shows the modem connection is successful and login works just fine, but for some reason ppp isn't doing its thing. On the server, I have tried every conceivable combination of setting in /etc/ppp/options /etc/ppp/options.ttyS1 and pap-secrets (which isn't needed with the -chap +pap login setting), but ppp still doesn't play nicely. The current options and option.ttyS1 configs are: [17:44 nemesis:/etc/ppp] # nc options debug kdebug 7 noauth crtscts lock modem asyncmap 200a0000 netmask 255.255.255.0 nodetach lcp-echo-interval 30 lcp-echo-failure 4 lcp-max-configure 60 lcp-restart 2 idle 600 noipx file /etc/ppp/filters proxyarp login ms-dns 192.168.7.15 ms-wins 192.168.7.15 [20:11 nemesis:/etc/ppp] # nc options.ttyS1 nemesis:ppp0 dnsaddr 192.168.7.15 192.168.7.65:192.168.7.66 I have tried both mgetty and faxgetty options in inittab, and faxgetty works fine to pass control to mgetty. The current inittab is: # modem getty (commented out) # mo:235:respawn:/usr/sbin/mgetty -s 38400 /dev/ttyS1 # fax getty (hylafax) (currently used) mo:35:respawn:/usr/lib/fax/faxgetty /dev/ttyS1 The mgetty.config is: [20:11 nemesis:/etc/ppp] # nc ../mgetty+sendfax/mgetty.config debug 4 fax-id 936 715 9339 speed 38400 port ttyS1 direct y toggle-dtr n The /var/log/mgetty.ttyS1 shows the following for each connect attempt: 09/09 17:45:11 yS1 mgetty: interim release 1.1.35-Feb22 09/09 17:45:11 yS1 WARNING: parent process not init(pid=1), but pid=12766 (/usr/lib/fax/faxgetty /dev/ttyS1) 09/09 17:45:11 yS1 check for lockfiles 09/09 17:45:11 yS1 huh? It's *our* lock file! 09/09 17:45:11 yS1 locking the line 09/09 17:45:11 yS1 we *have* the line! 09/09 17:45:11 yS1 waiting... 09/09 17:45:13 ##### data dev=ttyS1, pid=12795, caller='none', conn='DIRECT', name='', cmd='/bin/login', user='david' Now I don't know if this is normal or not. If I just use mgetty instead of faxgetty, then the connection fails stopping at the "waiting..." prompt above. I don't know if that is the root cause of the problem, but I'm not sure what to change to test it. All of this used to just work on my 10.0 and 10.3 boxes, but for some reason I am pulling my hair out with the 11.0 configuration. The adaptive answer part of the fax config is working fine as evidenced by the recognition of the "data" call received, but I am stumped as to why ppp doesn't give a good connection. I somebody has a working config for hylafax + dail-in ppp, I would sure appreciate some help with the configuration. The frustrating part is that I know it will work because I have done it before, but now something is amiss. Lastly, yes, /proc/sys/net/ipv4/ip_forward is set to 1 and I have tried the gettyargs of: #GettyArgs: "-h %l dx_%s" GettyArgs: "-r -b -s %s %l" and still nothing. Any help would be greatly appreciated. Thanks. -- David C. Rankin, J.D.,P.E. Rankin Law Firm, PLLC 510 Ochiltree Street Nacogdoches, Texas 75961 Telephone: (936) 715-9333 Facsimile: (936) 715-9339 www.rankinlawfirm.com -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org