https://bugzilla.novell.com/show_bug.cgi?id=469206 Summary: NIC without address fails to "come up" on boot, or with ifup, until "yast2 lan" is run, making automatic dsl connection hard Classification: openSUSE Product: openSUSE 11.1 Version: Final Platform: x86-64 OS/Version: openSUSE 11.1 Status: NEW Severity: Normal Priority: P5 - None Component: Network AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: andrew@acooke.org QAContact: qa@suse.de Found By: --- Created an attachment (id=267473) --> (https://bugzilla.novell.com/attachment.cgi?id=267473) The workaround script (cron job) User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-GB; rv:1.9.0.5) Gecko/2008121300 SUSE/3.0.5-1.1 Firefox/3.0.5 Hi, I realise I have a somewhat odd setup: I have a small network at home, and the server running OpenSuse acts as firewall, connecting to the 'net via a DSL modem (which is relatively new, used to have a simpler Cable model). This was working with 11.0, but when I upgraded to 11.1 (new install - have / and home on separate partitions, and a spare for installing upgrade, which makes things easy) it stopped working (although the DSL connection worked fine during install). The problem is that while I can configure everything just fine, and get everything to work, when I reboot, the NIC used to connect to the DSL (ppp over ethernet) does not "appear" (even though the boot.msg looks OK to me). This means (as far as I can tell from the error messages) that the dsl0 interface then fails (ie ppp fails). After experimenting, I have found that "yast2 lan show id=1" is sufficient to bring the NIC up correctly (after, I assume, "ifup eth2" occuring during boot). Once that is OK, "ifup dsl0" works. Since the DSL tends to die every day or two, I now have a script that runs as a 1min repeating cron that checks DSL and brings it back up. It also brings up the connection initially. I will include the script. (Oh, as an added bonus, knockd often jams at 100% cpu while playing around with the DSL interface - that's not a big issue for me, but you might see it while reproducing this and decide to raise an issue...) I hope that's clear enough. I guess you want more details about hardware etc. This is a fairly standard homebuilt box (Intel dual core, 4Gb memory). One NIC is on the mobo, the other is an old card I had lying around. I tried swapping the two (ie changing which is used for the internal network and which for the DSL) in case this was related to the card itself, but the problem persisted (in the process I ended up with eth1 and the2 instead of eth0 and eth1 - don't really understand where that numbering comes from). Here's the output from iconfig when everything is working: dsl0 Link encap:Point-to-Point Protocol inet addr:201.222.129.212 P-t-P:10.52.94.3 Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492 Metric:1 RX packets:1434 errors:0 dropped:0 overruns:0 frame:0 TX packets:1527 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:3 RX bytes:861965 (841.7 Kb) TX bytes:256898 (250.8 Kb) eth1 Link encap:Ethernet HWaddr 00:19:5B:2E:C2:F2 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:1532 errors:0 dropped:0 overruns:0 frame:0 TX packets:1575 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:899483 (878.4 Kb) TX bytes:291996 (285.1 Kb) Interrupt:22 Base address:0xb800 eth2 Link encap:Ethernet HWaddr 00:17:31:8E:5D:1F inet addr:10.2.0.1 Bcast:10.2.0.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:2631 errors:0 dropped:0 overruns:0 frame:0 TX packets:1889 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:426338 (416.3 Kb) TX bytes:267080 (260.8 Kb) Interrupt:252 Base address:0x4000 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:7096 errors:0 dropped:0 overruns:0 frame:0 TX packets:7096 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:2358528 (2.2 Mb) TX bytes:2358528 (2.2 Mb) And when things aren't working, eth1 and dsl0 are not present. Here's the output from "yast2 lan show" for the cards (incidentally, I am impressed with the amount of documentation/help that these scripts have - I was just blundering around and got things to work by reading "-help" output. This is really nice work (well apart from this bug I guess :o)) qp6 bin: yast2 lan show id=0 RTL8111/8168B PCI Express Gigabit Ethernet controller MAC : 00:17:31:8e:5d:1f Device Name: eth2 Started automatically at boot IP address: 10.2.0.1/24 qp6 bin: yast2 lan show id=1 VT6105/VT6106S [Rhine-III] MAC : 00:19:5b:2e:c2:f2 Device Name: eth1 Started automatically at boot I will include the ifcfgs. Here's provider0 (coying it so I can delete the pwd): qp6 network: cat providers/provider0 ASKPASSWORD='no' AUTODNS='yes' AUTO_RECONNECT='yes' DEFAULTROUTE='yes' DEMAND='no' DSLSUPPORTED='yes' IDLETIME='600' IPADDR='' ISDNSUPPORTED='no' MODEMSUPPORTED='no' MODIFYDNS='no' MODIFYIP='yes' PASSWORD='TCHILE' PHONE='' PROVIDER='speedy' REMOTE_IPADDR='' USERNAME='xxx' As I think I have already said, I will also include my ad-hoc script that I use as a workaround. Feel free to ask for more info. I have worked with you on a previous bug and, although it was a bit slow, we got there eventually. PS If it's relevant - I'm a programmer whose used Linux for years, but I'm out of my depth when it gets past running "ifup"... Reproducible: Always Steps to Reproduce: 1. Reboot Actual Results: No eth1 or dsl0 in "ifconfig" (ie no network connection). Expected Results: eth1 and dsl0 present in "ifconfig" (ie network connection) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.