Re: Assorted arm-buster problems - network configuration
On Thursday 04 July 2019 16:05:09 tomas@tuxteam.de wrote:
> On Thu, Jul 04, 2019 at 08:56:45PM +0100, Tixy wrote:
> > On Thu, 2019-07-04 at 20:01 +0100, Brian wrote:
> > > On Thu 04 Jul 2019 at 19:18:13 +0300, Reco wrote:
> > >
> > > [...]
> > >
> > > > I'd also consider exterminating avahi with extreme prejudice,
> > > > i.e. 'apt
> > > > purge avahi-daemon'. Really simplifies things. Not installing
> > > > this software in the first place works even better.
> > >
> > > Gene Heskett can follow this advice if he wishes. It is to be
> > > hoped that every other user ignores it.
> >
> > Why? It's advice I decided for myself 10 or more years ago after
> > seeing constant reports of zeroconf bugs in various OSes and kit,
> > and realising that sort of thing was also running on my Linux
> > machines. The whole idea of automagically setting up networks just
> > sounds like a problem and security hole waiting to happen. So I
> > decided to nuke it from orbit, it was the only safe thing to do.
>
> As always, all generalizations suck. Some do avahi, others don't (full
> disclosure: I am in the "don't" camp, as many may have guessed :-)
>
> Still, I dislike such statements as "exterminate". But YMMV.
>
> Cheers
> -- t
In this case, it made zero difference. It. nsswtch, dhcpd and all its ilk
have been purged, but the default routes remain stubbornly, even over
powerdown reboots, at 169.254.0.0/16. I next will grep the whole system
looking for it and any file containing it will get edited to replace it
with MY 192.168.nn/24 domain. If that doesn't fix it, it gets nuked.
Since It nothing but a 64GB u-sd card I can rewrite in 10 minutes, I
officially don't care if I destroy its ability to boot. It also has a
duff kernel, RealtimePi having been watched as it spent an hour building
a 4.4.114 rt kernel, but then reused the 4.19.50 kernel in the original
image's zip. So I wind up with an identicaly functioning 2019-06-20
buster image for nominally 4 hours work by the pi 3b.
I do have a good stretch version but with an earlier rt kernel, which
works very well except for the nominal 2 frames a second video speeds.
latency-test reports 28 u-secs of jitter in the 1 kilohertz servo
thread, compared to a jessie builds 70+ microseconds.
Buster however is supposed to have faster video, but w/o a network I
can't install the test tools to measure it, this is the main pusher to
make buster work. And I can't measure the video speeds without a
network.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>
Reply to: