From: Florent C. <Flo...@un...> - 2003-01-22 22:05:28
|
Hi I gave a try to clustermatic today, on a test cluster made of 8 nodes with dual pentium 3 processors, serverwork chipset with integrated eepro100 , + one eepro on the master, myrinet 2000 boards and switch. Either with fast ethernet or myrinet the nodes hang on the boot.img fetch after they got an IP from RARP. I tried the mcastbcast hack, crossover cables instead of our Cisco switch, and boot over myrinet (with different MACs ) : all the same, nothing happens and tcpdump does not show anything after the RARP resolution. I chose addresses in the 192.168.33 range in order not to interfer with the campus here. Did I forget something in the redhat 8.0 security settings ? I set no firewall and started all services... Please help thanks in advance -- Florent Calvayrac | Tel : 02 43 83 26 26 Laboratoire de Physique de l'Etat Condense | Fax : 02 43 83 35 18 UMR-CNRS 6087 | http://www.univ-lemans.fr/~fcalvay Universite du Maine-Faculte des Sciences | 72085 Le Mans Cedex 9 |