Re: DS10 Installation problems
On Wed 25 Oct 2000, Marc Dubrowski wrote:
> During the last month, the tape-drive cable was finally installed on the
> machine, still with tru64. After that, when I wanted to boot Debian, I
> discovered after an hour that the OS would'nt discover the second hard drive
> as long as no tape was in the tape driver. Can someone explain me why ?
I vaguely recall reading something like this, it was some bug or so
in the scsi host adapter driver. What adapter is it? maybe not probing
multiple LUNs will also help (kernel config option).
> Second problem is with the console a vt420 terminal on the serial port.
>
> When I installed th OS, it worked fine. Since the tape Installation (I guess,
> since nothing else has changed) the boot message freeses on the console when
> the portmap daemon is booting. (I removed the portmap daemon from the boot
> process, and the freeze occured after the network interfaces are setup).
Sounds more like a name resolving timeout issue. Anything in the way
of DNS servers changed, and not updated in the system?
> More, the text on the terminal seems to shake slowly, which I have never seen
> before.
> I'm not used with working with terminals, as my experience is mostly on
> smaller machines, But I installed a AS1000 without a problem. Can someone
> point me to an explanation ?
The shaking of text on a vt420 isn't something linux can influence :-)
You usually see this when the terminal is next to another device (e.g.
another terminal, monitor, or power supply of some computer), and the
(magnetic) interference from that other device is deflecting the
electron beam.
> Note that at that point, I am able to log in the machine from the network.
> Except it is very slow before the login prompt. I suspect it may be because of
> the autonegociation between the computer and the switch, which I'll fix
> tonight. My two NICS are from DEC, model tulip.
No, this is probably also due to a name resolving timeout. You
connect from another system, and the debian system wants to know
what the name of that other system is (e.g. for logging and/or
tcp wrappers), and that fails. Check your nsswitch.conf, resolv.conf,
/etc/hosts, ...
> And, finally, I'd like to configure that server as a mail gateway for all my
> organisation, but without it habing any users. All incoming messages should be
> redirected to our internal mail server, but it would also relay all mail for
> the internet coming form our internal mailserver to the internet.
>
> If I am not wrong, the internal mail server would be the MAIL_HUB, and the
> server would be the smart host. Am I right ? Can someone poit me to clear doc
> for that ?
It's not quite like that. The server is not the smart host, but the
smart user host. It's something that is very easily done with exim
(the standard mailer with debian), see the exim FAQ (either in the
exim-doc package or on the www.exim.org website).
Paul Slootman
--
home: paul@wurtel.demon.nl http://www.wurtel.demon.nl/
work: paul@murphy.nl http://www.murphy.nl/
debian: paul@debian.org http://www.debian.org/
isdn4linux: paul@isdn4linux.org http://www.isdn4linux.org/
Reply to: