Re: [cobalt-users] The ongoing saga of the dying RaQ3

[Jeff Lasman <jblists@xxxxxxxxxxxxx>]

Chae wrote:

> If you remember I have the RaQ3 that keeps dying on me - well not dying the
> servers still up but no-one can access it via http, ftp or ssh, logcheck

Can you ping it?

> After a reboot's done everything is back to normal,

This is important information... keep reading <smile>...

> 
> A lot of you suggested that it could be a network connection problem, loose
> connection, faulty network card etc.

RaQs don't have network cards.  They have a built-in network
connection.  So the only thing that could wiggle loose would be the
actual plugged-in eithernet wire.  Which isn't happening if everythings
back to normal during a reboot... unless of course the machine is moved
while rebooted.

> "...that's an internal sendmail error. 127.0.0.1 is the loopback ip. it's
> not a "network" per say. just a virtual ip that daemons use to refer back
> to themselves.

While I might not use exactly the same terminology, they're right.

> if any program has problems connecting to 127.0.0.1 it is definitely a
> problem with the daemon/daemon configuration."

That's true as well.

> So I'm back to square one again or I have a colo who doesn't want to open
> up the RaQ3 to check the network card etc or doesn't want to go to the
> bother of doing it :(

As I said, there's no network card to check.

> Anyone any suggestions on how I can pin-point the problem myself - can't
> access this server as it's half-way round the world.

Have you made sure you're not hacked?  At the very least, get a copy of
the latest "chkrootkit" (http://www.chkrootkit.org/), install it, and
run it, and let us know if you get any positives. Don't faint if you do
<grin>... there are such things as false positives.

Jeff
-- 
Jeff Lasman <jblists@xxxxxxxxxxxxx>
Linux and Cobalt/Sun/RaQ Consulting
nobaloney.net, P. O. Box 52672, Riverside, CA  92517
voice: +1 909 778-9980  *  fax: +1 909 548-9484