[HECnet] Introducing myself... and my little network

[Jordi Guillaumes i Pons]

El 10/06/2012, a les 0:33, Peter Lothberg va escriure:



Now I've got another problem. The TOPS-10 node goes yo-yo:
(...)
$

It's outside tops10, I guess, and I'm completely lost there. If my
braincells starts to work I will rember how to log all the DECnet
packets on the -10 side.

--P



Its fixed now. I'm running the PDP-10's in a virtualbox VM, and I'm using a virtualbox feature that allows to cap the %CPU of the host CPU the VM can use. If I cap it to under 40% the virtual ethernet devices begin to drop packets and the yo-yoing begins. Curiously, TOPS-20 is more resilient and, although I see its ethernet dropping as much packets as the TOPS-10 one it does not drop itself from the network...

KLH10 does not seem to have an idle loop detection like SIMH does. The Panda TOPS-20 uses some sort of virtual device to make the host aware if it is idling, but the regular TOPS-10 monitor does not, so it's using the 100% of the CPU time of the host virtualbox machine. So if I don't cap it it ends topping one of the cores of the "real" host machine. And the it gets hot :) 


Jordi Guillaumes i Pons
jg at jordi.guillaumes.name
HECnet: BITXOV::JGUILLAUMES