dhcpcd-discuss

The original poster of the netctl issue did some more experiments and
found out that a very similar issue occurs when the device disappears.
With a USB network adapter, this can happen when the adapter is
unplugged. I guess that may be easier to simulate in a VM than a
suspend.

It is pretty warm at my place, so I don't spend too much time behind a
computer, but here is another hypothesis on why some people are seeing
their systemd services time out and killing dhcpcd forcefully. Could
it be that in some cases dhcpcd spans multiple system processes and
does not kill all of them in case of an error? If one process reports
'dhcpcd exited', but other processes remain, then systemd is right to
infer that the service as a whole has not exited.

Thanks,
- Jouke


On Tue, Aug 11, 2020 at 1:46 AM Neal P. Murphy
<neal.p.murphy@xxxxxxxxxxxx> wrote:
>
> > So, the actual bug report is that dhcpcd is "inactive".
> > The errors reported are just errors (which I should try and fix if I can
> > reproduce it on a Linux VM - how do I suspend and resume that?).
>
> You should be able to manually tell the linux instance to hibernate. When you next boot it, it should 'resume'. You may have to ensure that you have enough swap space defined so everything can be stored.
>
> N
>