[systemd-devel] systemd-networkd not discovering all devices at bootup, and thus no network is configured

[Keller, Jacob E]

On Thu, 2015-02-12 at 14:23 +0100, Tom Gundersen wrote:
> On Thu, Feb 12, 2015 at 11:45 AM, Lennart Poettering
> <lennart at poettering.net> wrote:
> > On Wed, 11.02.15 22:28, Keller, Jacob E (jacob.e.keller at intel.com) wrote:
> >
> >> On Wed, 2015-02-11 at 21:46 +0100, Lennart Poettering wrote:
> >> > On Wed, 11.02.15 20:36, Keller, Jacob E (jacob.e.keller at intel.com) wrote:
> >> >
> >> > > > > Yes. I believe that possibly the netlink messages are occurring too
> >> > > > > early before networkd has started. Thus, when I restart the service
> >> > > > > after startup it works fine.
> >> > > >
> >> > > > Well, the first thing after subscribing to links coming/going that
> >> > > > networkd does is query the kernel for the list of devices it has. This
> >> > > > means that it should always get all links, regardless when it is started.
> >> > >
> >> > > What about network device renames? Devices are renamed after they
> >> > > appear. Is it also possible that networkd is started too early and thus
> >> > > the devices haven't actually appeared yet? But.. then it should get the
> >> > > message once it does appear? (assuming here that it continues listening
> >> > > to rtnl messages about links)
> >> >
> >> > networkd should get all netlink msgs about all of this just fine, and
> >> > be able to track renames and everything.
> >> >
> >> > > How difficult would it be to create an RPM?
> >> >
> >> > Well, you can get the spec file from fedpkg and then use it on a
> >> > tarball from systemd git made with "make dist". But you probably need
> >> > to make a number of updates to the .spec file. Unless you are an RPM
> >> > guru I wouldn't bother.
> >> >
> >> > Lennart
> >> >
> >>
> >> Ok. I got the upstream systemd installed. Now I get a very different
> >> behavior, which I am unsure of.
> >>
> >> I cut off the last message as it repeated some odd ten thousand times or
> >> so... Presumably retry/failing...
> >>
> >> At any rate, it appears possible that some configuration is wrong, or
> >> similar? Any suggestions/thoughts now?
> >>
> >> Feb 11 14:21:47 jekeller-copperpass systemd-networkd[1068]: enp4s0f0        : flags change: +UP
> >> Feb 11 14:21:47 jekeller-copperpass systemd-networkd[1068]: enp4s0f0        : enp4s0f0        : could not bring up interface: Invalid argument
> >
> > OK so this is already weird.
> 
> This should now be fixed in git (5c3072eab6d2e11d89452987b017541d4654ac05).
> 
> >> Feb 11 14:21:52 jekeller-copperpass systemd-networkd[1068]: DHCP CLIENT (0xc7096673): could not receive message from raw socket: Invalid argument
> >
> > Hmm, this one might actually be caused by
> > d6d810fbf8071f8510450dbacd1d083f37603656. WIll revert that, please
> > retest then!
> >
> > Lennart
> >
> > --
> > Lennart Poettering, Red Hat
> > _______________________________________________
> > systemd-devel mailing list
> > systemd-devel at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Ok, I'll try with latest Git first, and if it still fails I'll try
reverting the patch Lennart suggested.

Thanks for the help!

Regards,
Jake