Re: tzdata problems etc
On Mon, Jul 08, 2013 at 11:25:59PM +0200, Petter Reinholdtsen wrote:
> [Giorgio Pioda]
> > Hi Petter,
> Hi. :)
> >> Which installation choices did you select?
> > I used the NETINST iso image burned on a CD.
> I mean which language and profile choices did you select during
> installation? Did you change the desktop selection on the boot
Italian language, Swiss Italian location, Swiss Frech Keyboard,
like allways (a little exotic, I know, but we live here)
> > Long output in kdm.log mainly a:
> > kdmgreet (4141)/kdecore (K*TimeZone*) complaining that ktimezoned
> > D-Buss call failed
> Hm, never seen that one myself. Are you sure it is relevant and not
> happening also with successful installs? Anything interesting in
> /var/log/auth.log from the failing login attempt?
I have to check at work, later this morning
> >> It should list http://webcache:3128 as the proxy, and the setting is
> >> done by debian-edu-pxeinstall. This do not happen when I test it, but
> >> I have seen it some times. Can you tell us more about how you
> >> installed the main-server in question? What was the network
> >> environment during installation and first boot?
> > I fixed it manually putting the correct IP, now squid gets populated
> That is a workaround, but I would love to figure out the root cause.
> The PXE setup inherit the proxy setting from the host, and if for some
> reason this is wrong, the PXE setup will be wrong too.
> > Not that bad:
> > -error: ./doc: RElease manual have 9 FIXME....
> > -error: ./nagios: Nagios count NUMSVCCRIT is not zero but 1 (two lines)
> Probably the DNS test reporting that it can't look up
> www.skolelinux.org. Did the machine have Internet connectivity and
> access to do DNS lookups externally during boot?
> > -error: ./webcache: WPAD file 'http://wpad/wpad.dat' is missing HTTP
> > proxy info.
> This is the indicator of the proxy problem. I am not sure how to
> trigger it. Perhaps you can help by telling more about the network
> environment where you installed the main server? I suspect it is
> relevant, as I do not get this bug myself when I test.
I was reading the thread that taffit has started tonight... Maybe related?
Anyway, I have a 10.164.88.0/23 segment given by the Telephon provider. All the
schools are inside this single large 10.0.0.0 network. I have a
dchp/dns running at 10.164.88.2, gateway 10.164.88.1. DHCP
passes also a secundary dns outside the /23 net (10.164.6.2)
For the NETINSTALL I leave the automatic preseed work (thus dchp).
Then after reboot I change the subnet with the provided script.
I also have a apt-cacher-ng which is served by a public IP, not in the
10.* lan but I didn't use it and I cannot figure out how it could be
I don't see any relevant point here.
Giorgio Pioda - Sysadmin SPSE-Tenero
Cell +41 79 629 20 63
Uff. +41 91 735 62 48