Re: Sparc Classic: ethernet hardware issue
Will Roberts <[email protected]> Fri, 23 May 2003 16:34:00 -0400
| Newsgroups | gmane.linux.suse.sparc |
|---|---|
| Message-ID | <[email protected]> |
At 10:35 AM 5/23/03, David List <[email protected]> wrote: >On Fri, 23 May 2003, Will Roberts wrote: > > >I am trying to install SuSE linux 7.3 for Sparc onto an old Sparc Classic. > >All is well except I discovered that the Classic has a bad on-board > >ethernet port. This causes both the obvious problem (no network connection) > >and the side effect of: > > > > INIT: Id "3" respawning too fast: disabled for 5 minutes > > INIT: Id "6" respawning too fast: disabled for 5 minutes > > INIT: Id "2" respawning too fast: disabled for 5 minutes > > INIT: Id "4" respawning too fast: disabled for 5 minutes > > etc. >...<snip>... > >You are not telling whether that is the case, but I bet you are running >the system headless, right? If I'm right, your problem has probably got >nothing to do with the networking device. I had exactly the same problem >right after installing on my Ultra1, which I run headless. > >The solution (well, the symptom-treatment, to be exact) was to >outcomment the 'mingetty' lines in /etc/inittab. I don't know how to >*really* cure the problem. I have seen it mentioned somewhere that this >might be about a bug in 'init'. Remember to remove the outcommenting >from the lines in /etc/inittab, when you connect monitor and keyboard to >the system again. > >Best regards, >David List David, thank you very much. I received several similar responses from other SuSE-sparc users and have commented the minigetty lines as suggested And, yes, it solves the respawning problem. I am still struggling with the other problem related to the ethernet port. I have managed to configure things so that SuSE recognizes the 501-2981 SBus ethernet card as eth1 and complains if I boot the system with that card removed. When the Sparc powers up, the ethernet link light on the connected hub comes on as would be expected. The routing table looks OK: SuSE-Q:/ # route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 172.16.2.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo 0.0.0.0 172.16.2.100 0.0.0.0 UG 0 0 0 eth1 which is exactly like the routing table in use on a well-functioning SuSE 7.0 (i386) system running on a Pentium-based machine the same LAN. The only obvious difference is that I have to run "route -n" with the "-n" flag. This forces the routing table to be displayed with numeric addresses. Otherwise "route" on the Sparc hangs after displaying only the column headings. My /etc/hosts file is populated with appropriate IPs and names for localhost, the machine itself, and the network gateway machine. It was suggested that I reconfigure eth0 (the broken port) with a non-routable unused network address because there may be other places in the SuSE distribution that makes reference to eth0 and causes the network functions to hang. I have not done this yet, but I will give it a try. Regards, Will The Old Bear