RH Fedora network interfaces

Chris redhat-list at dotcomdesigners.com
Thu Feb 19 08:39:49 UTC 2004


Just updating my own message in case someone reading this has ANY ideas on what
else to try.  See below.


> > Chris, I don't think it is so much an initialization at start-up that's
> > failing, I'm thinking that on a warm boot as eth0 and 1 shut down,
> > somehow the card is left in a jammed state. A warm boot does not remove
> > power from the PCI bus (yes, even if they're on-board they're on the PCI
> > bus). A cold boot does, hence in effect re-setting the card back to
> > blank slate so to speak.
> >
> > I'm thinking the module for the card is not shutting the card back to
> > blank slate mode on a warm boot.
> >
> > If you bring them down manually, then up again, what happens?
> >
> > Or more basic level, what about removing the module and re-installing it?
> >
> > Does Intel provide a Linux driver for that card?
> >
> > Those are the things I would look at.
> >
> > Regards,
> > Ed.
>
> I can bring up and down both interfaces after a cold boot and everything works
> great.  But if I do a warm boot, the problem comes back to life - RH9 / Fedora
> Core 1 will not see network ports.
>
> I have downloaded the latest Intel drivers for this card, Pro/100 82559, and
> when I 'make install' it goes okay but the last line says:
>
> depmod: *** Unresolved symbols in
> /lib/modules/2.4.22-1.2149.nptlsmp/kernel/drivers/net/e100.o
>
> Is this a cause for concern?  (I am not physically in front of the server as I
> write this so I didn't want to remotely reboot the sucker in case it doesn't
> come back up and I have to drive to the datacenter again).
>
> I did some digging on the error message above and in somewhat similar cases
> (problems with getting network ports to work properly), I can also add that my
> /etc/modules.conf has:
>
> alias eth0 e100
> alias eth1 e100
>
> which appears to be correct.  The only other thing someone else had posted on
a
> vaguely related subject was that in his case some of the bootup scripts were
> trying to access network ports ahead of 'ifconfig eth0 up' which was running
> much later on in the bootup process.  Could this be a similar problem?  Where
> can I look for the order this gets loaded in and where can I change it so that
> 'ifconfig eth0 up' runs later in the process?  (if that is a problem, that is)
>
> Thanks,
>
> Chris

I just came back from the datacenter.  Removed the old e100 drivers (rmmod
e100), compiled the latest Intel Pro/100 82559 drivers I could find
(e100-2.3.27.tar.gz), installed them (insmod e100), did modprobe and all that,
and it "sees" the ports.  I have full connectivity.  Do a full shutdown, boot up
again, everything works PERFECTLY.  Cool.  Do a warm-reboot - and we're back to
square one - Fedora comes up with the same exact message, e100 device cannot
find eth0 / eth1, and manually trying to bring these interfaces up with ifup
eth0 / ifup eth1 results in... nothing (same error message as during bootup
about missing e100 devices).

Ideas?  Suggestions?

Chris






More information about the redhat-list mailing list