I have updated to baselayout-1.12.0-r1 & ppp-2.4.3-r15 & rp-pppoe-3.8 ; my kernel version is 2.6.14 . I can connect to my ISP (Sympatico, Toronto) using the ADSL module in /etc/conf.d/net , but cannot connect from the PPP module. I have copied what seem to be the relevant lines from net.example : config_eth0=( "192.168.0.2 netmask 255.255.255.0 broadcast 192.168.0.255" ) config_ppp0=( "ppp" ) link_ppp0="eth0" plugins_ppp0=( "pppoe" ) username_ppp0='user@isp' pppd_ppp0=( "updetach" "noauth" "defaultroute" "usepeerdns" "persist" ) and have made the required symlink in /etc/init.d : net.ppp0 -> net.lo . The following lines appear in syslog for the 2 methods (without/with Rp): using PPP alone : fails 16:08:32 kernel: ReiserFS: hda9: Using r5 hash to sort names 16:08:36 kernel: skge eth0: enabling interface 16:08:38 kernel: skge eth0: Link is up at 10 Mbps, half duplex, flow control none 16:08:41 ntpd[4137]: ntpd 4.2.0a@1.1190-r Sat Feb 4 09:05:38 EST 2006 (1) 16:08:41 ntpd[4137]: precision = 1.000 usec 16:08:41 ntpd[4137]: Listening on interface wildcard, 0.0.0.0#123 16:08:41 ntpd[4137]: Listening on interface eth0, 192.168.0.2#123 16:08:41 ntpd[4137]: Listening on interface lo, 127.0.0.1#123 16:08:41 ntpd[4137]: kernel time sync status 0040 using Rp-pppoe : succeeds : additional lines marked : 16:13:03 kernel: ReiserFS: hda9: Using r5 hash to sort names 16:13:07 kernel: skge eth0: enabling interface >>16:13:07 pppd[3932]: pppd 2.4.3 started by root, uid 0 >>16:13:07 pppd[3932]: Using interface ppp0 >>16:13:07 pppd[3932]: Connect: ppp0 <--> /dev/pts/0 16:13:09 kernel: skge eth0: Link is up at 10 Mbps, half duplex, flow control none >>16:13:12 pppoe[3933]: PADS: Service-Name: '' >>16:13:12 pppoe[3933]: PPP session is 6907 (0x1afb) >>16:13:16 pppd[3932]: PAP authentication succeeded >>16:13:16 kernel: PPP BSD Compression module registered >>16:13:16 pppd[3932]: local IP address 70.49.170.21 >>16:13:16 pppd[3932]: remote IP address 64.230.197.176 >>16:13:16 pppd[4070]: Can't open options file /etc/ppp/peers/noauth: No such file or directory 16:13:20 ntpd[4310]: ntpd 4.2.0a@1.1190-r Sat Feb 4 09:05:38 EST 2006 (1) 16:13:20 ntpd[4310]: precision = 1.000 usec 16:13:20 ntpd[4310]: Listening on interface wildcard, 0.0.0.0#123 16:13:20 ntpd[4310]: Listening on interface lo, 127.0.0.1#123 >>16:13:20 ntpd[4310]: Listening on interface ppp0, 70.49.170.21#123 16:13:20 ntpd[4310]: kernel time sync status 0040 I have searched bugs & Forum without finding any sign of this problem & have received no response to a request for advice on the User mailing-list. The User Manual still recommends the Rp approach. The Baselayout & PPP versions above are "testing", so hopefully this will be useful as a test result of the new method.
Did you add net.ppp0 to the autostart (rc-update add net.ppp0 default) scripts? It seems pppd is not even started... Your kernel has pppoe support? (not required for rp-pppoe, but for this config!) Your config looks ok otherwise, a very similar one works flawlessly for me. I use config_eth0=( "null" ), since that interface is only used for pppoe and doesn't need any addresses, but that all of the difference. You might also want to try to add "mtu 1492" and "mru 1492" to your pppd options, as that is what most people recommend (ethernet mtu - pppoe header size)
please add "debug" to pppd_ppp0 and post here the resulted logs.
(In reply to comment #1) > It seems pppd is not even started... Yes, that's very much how it looks to me. I don't know why. > Did you add net.ppp0 to the autostart scripts? I already have net.eth0 there & both are symlinks to net.lo , so it shouldn't make any difference. > Your kernel has pppoe support? Yes.
(In reply to comment #2) > please add "debug" to pppd_ppp0 and post here the resulted logs. I have done that & there is no change to the syslog output or boot messages. The output from /var/log/debug is below, but it doesn't seem to have any lines referring to PPP : Jun 8 21:59:10 localhost kernel: On node 0 totalpages: 261952 Jun 8 21:59:10 localhost kernel: DMA zone: 4096 pages, LIFO batch:1 Jun 8 21:59:10 localhost kernel: Normal zone: 225280 pages, LIFO batch:31 Jun 8 21:59:10 localhost kernel: HighMem zone: 32576 pages, LIFO batch:15 Jun 8 21:59:10 localhost kernel: ACPI: RSDP (v000 ACPIAM ) @ 0x000f91e0 Jun 8 21:59:10 localhost kernel: ACPI: RSDT (v001 A M I OEMRSDT 0x05000407 MSFT 0x00000097) @ 0x3ff40000 Jun 8 21:59:10 localhost kernel: ACPI: FADT (v002 A M I OEMFACP 0x05000407 MSFT 0x00000097) @ 0x3ff40200 Jun 8 21:59:10 localhost kernel: ACPI: MADT (v001 A M I OEMAPIC 0x05000407 MSFT 0x00000097) @ 0x3ff40300 Jun 8 21:59:10 localhost kernel: ACPI: OEMB (v001 A M I OEMBIOS 0x05000407 MSFT 0x00000097) @ 0x3ff50040 Jun 8 21:59:10 localhost kernel: ACPI: DSDT (v001 A0008 A0008205 0x00000205 INTL 0x02002026) @ 0x00000000 Jun 8 21:59:10 localhost kernel: ACPI: Local APIC address 0xfee00000 Jun 8 21:59:10 localhost kernel: ACPI: IRQ0 used by override. Jun 8 21:59:10 localhost kernel: ACPI: IRQ2 used by override. Jun 8 21:59:10 localhost kernel: ACPI: IRQ9 used by override. Jun 8 21:59:10 localhost kernel: mapped APIC to ffffd000 (fee00000) Jun 8 21:59:10 localhost kernel: mapped IOAPIC to ffffc000 (fec00000) Jun 8 21:59:10 localhost kernel: CPU: After generic identify, caps: 0383fbff c1c3fbff 00000000 00000000 00000000 00000\ 000 00000000 Jun 8 21:59:10 localhost kernel: CPU: After vendor identify, caps: 0383fbff c1c3fbff 00000000 00000000 00000000 000000\ 00 00000000 Jun 8 21:59:10 localhost kernel: CPU: After all inits, caps: 0383fbff c1c3fbff 00000000 00000020 00000000 00000000 000\ 00000 Jun 8 21:59:10 localhost kernel: Boot video device is 0000:01:00.0 Jun 8 21:59:10 localhost kernel: ACPI: PCI Interrupt Routing Table [\_SB_.PCI0._PRT] Jun 8 21:59:10 localhost kernel: PCI: Setting latency timer of device 0000:00:01.0 to 64 Jun 8 21:59:10 localhost kernel: Probing IDE interface ide0... Jun 8 21:59:10 localhost kernel: Probing IDE interface ide1... Jun 8 21:59:10 localhost kernel: libata version 1.12 loaded. In case it's important, the relevant section of /usr/src/linux/.config is : # Wan interfaces # # CONFIG_WAN is not set # CONFIG_FDDI is not set # CONFIG_HIPPI is not set # CONFIG_PLIP is not set CONFIG_PPP=y # CONFIG_PPP_MULTILINK is not set # CONFIG_PPP_FILTER is not set CONFIG_PPP_ASYNC=y # CONFIG_PPP_SYNC_TTY is not set CONFIG_PPP_DEFLATE=y CONFIG_PPP_BSDCOMP=m CONFIG_PPPOE=y # CONFIG_SLIP is not set # CONFIG_NET_FC is not set # CONFIG_SHAPER is not set # CONFIG_NETCONSOLE is not set # CONFIG_NETPOLL is not set # CONFIG_NET_POLL_CONTROLLER is not set It looks as if PPP is not starting, but I have no idea why.
That is strange. I see you are using "updetach"... When you start net.ppp0, it just freeze waiting for the link to come up, right?
(In reply to comment #5) > That is strange. I see you are using "updetach"... > When you start net.ppp0, it just freeze waiting for the link to come up? I've solved the problem, though I'm not quite sure what should be happening. Using Rp-pppoe , I need 'net.eth0' in /etc/runlevels/default , but to use PPP I need 'net.ppp0' there instead. I was confused -- and still don't understand what's going on -- by the fact that both are symlinks for 'net.lo', which has already started before Runlevel 3 is entered. How can they have different effects like this ? Also, I tried removing all the items in pppd_ppp0 except 'debug' & the connection didn't go through, so something in there is essential. Do you have any advice what ? I can try removing each & see what happens. You do need to add to /etc/conf.d/net.example a line to tell users to make sure 'net.ppp0' is in their default runlevel: that's a step which will not be clear to everyone (it is/wasn't to me). Thanks a lot for your attention. If you could answer my 2 questions above, it would help me understand the process a lot better.
(In reply to comment #6) Some experimentation with flags in pppd_ppp0 in /etc/conf.d/net : 'updetach' is needed to prevent Ntpd trying to start before ADSL & stalling; 'defaultroute' is needed: the connection is made to the ISP, but Lynx can't connect to Internet sites (tested twice); the other flags I listed earlier don't seem to make a difference. All seems to be working now & thanks again for looking at this, but I do suggest net.example needs a couple more lines of clarification.
(In reply to comment #6) > Also, I tried removing all the items in pppd_ppp0 except 'debug' > & the connection didn't go through, so something in there is essential. > Do you have any advice what ? I can try removing each & see what happens. Well, usually you'll need "noauth" "defaultroute" "usepeerdns". net.example already advise you to read in pppd's manual page and that is all you need to know. > You do need to add to /etc/conf.d/net.example > a line to tell users to make sure 'net.ppp0' is in their default runlevel: > that's a step which will not be clear to everyone (it is/wasn't to me). This is an obvious thing to do. Why would you need to make that net.ppp0 symlink and set config_ppp0 if you don't start net.ppp0 (ppp0 depends on eth0, not the other way around)? You're the only one till now who didn't understand it and, believe me, there are many peeps that already use ppp net module. rp-pppoe module is a broken way of setting up a PPPoE link from Gentoo's pov, but many noobs find it easier to use. Since Gentoo is all about choices ...
