dhcpcd-discuss

I ran it with the new patch.  It still got a failure (system call 334).
 To speed up the debug cycle a bit I went ahead and added another entry for
that syscall (NR_send it looked like...).    And it started working.

I have attached a patch which includes your patch plus my
additional NR_send entry.    I rebuilt dhcpcd 9.3.1 clean with this patch
applied and it worked.

I didn't attach strace or log dumps but if you would like those, I can send
them.

On Thu, Oct 15, 2020 at 10:01 AM Roy Marples <roy@xxxxxxxxxxxx> wrote:

> On 15/10/2020 15:47, Tom Armistead wrote:
> > I think I may have made a mistake regarding strace.dump3.   I think when
> I added
> > the second patch, I mistakenly wound up not applying any patch at all.
>  So, if
> > strace.dump3 doesn't make sense, that is probably due to an error on my
> part..
> >
> > I have applied your debugging patch and ran it.   The output when I
> start dhcpcd
> > is attached as dhcpcd.debug.log.    It looks like it does have the
> syscall
> > message that you were hoping for.
> >
> > In case you might need it, I also did an strace with your new debug
> patch
> > applied and that output is attached as strace.dump4.
>
> And what is *really* helpful is that strace was intelligent enough to
> translate
> the syscall number into __NR_time for me :)
>
> Attached is a replacement patch which now allows this.
> Lets see how far we get now!
>
> Roy
>

Attachment: tom.dhcpcd.patch
Description: Binary data