On 09/21/2018 02:57 PM, Ken Moffat wrote:
On Fri, Sep 21, 2018 at 12:20:01PM -0500, Bruce Dubbs wrote:
On 09/21/2018 12:28 AM, Bruce Dubbs wrote:
On 09/21/2018 12:02 AM, Ken Moffat wrote:
The build seemed to go ok (I haven't looked at the test results
beyond noticing that I'm still seeing failures in elfutils although
a run on a completed 8.3 system was ok - I had guessed maybe it
needed more static libs, but apparently not)
But apart from the weird build failure with nfs-utils (see blfs-dev)
my attempt to successfully boot failed - device eth0 not present.
I don't know about eudev causing eth0 to go missing. Take a look at
/sys/class/net/ and see if the nic was detected. Of course the name
should be handled by
/etc/udev/rules.d/70-persistent-net.rules
For System V, I have /sys/class/net/eth0 and was able to ping 8.8.8.8 just
fine.
thanks for looking at this. Unfortunately it was luser error - I'd
copied an entry in grub.cfg, changed the kernel version but failed
to change rootfs=. Often I'm already running the kernel I plan to
use (on the host system), but I figured the differences between
4.18.6 and 4.18.9 were small enough [¹].
Yup. I do that too.
Chrooted to it a few minutes ago, the persistent net rules looked
fine. Went to look at the syslog in chroot, not present. It's been
a bad week, didn't immediately think of the obvious. Booted it, no
modules loaded. Checked that there were modules in /lib/modules,
tried to modprobe atl1c, failed, module not found. That's when I
realised that I was running 4.18.9 but the modules I was looking at
were for 4.18.6.
Sorry for the noise.
LOL. Been there, done that.
-- Bruce
--
http://lists.linuxfromscratch.org/listinfo/lfs-dev
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
